An Assessment of the Effect of 
Devolution on the Healthcare Service 
Delivery in ASAL Counties of Kenya
Wilkista Lore and Susan Thuo
Kenya Institute for Public Policy 
Research and Analysis
KIPPRA Discussion Paper No. 316
2023
 
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
KIPPRA in Brief
The Kenya Institute for Public Policy Research and Analysis (KIPPRA) is an 
autonomous institute whose primary mission is to conduct public policy research 
leading to policy advice. KIPPRA’s mission is to produce consistently high-quality 
analysis of key issues of public policy and to contribute to the achievement 
of national long-term development objectives by positively influencing the 
decision-making process. These goals are met through effective dissemination 
of recommendations resulting from analysis and by training policy analysts in 
the public sector. KIPPRA therefore produces a body of well-researched and 
documented information on public policy, and in the process assists in formulating 
long-term strategic perspectives. KIPPRA serves as a centralized source from 
which the Government and the private sector may obtain information and advice 
on public policy issues.
Published 2023
© Kenya Institute for Public Policy Research and Analysis
Bishops Garden Towers, Bishops Road
PO Box 56445-00200 Nairobi, Kenya
tel: +254 20 2719933/4; fax: +254 20 2719951
email: admin@kippra.or.ke
website: http://www.kippra.org
ISBN 978 9914 738 41 4
The Discussion Paper Series disseminates results and reflections from ongoing 
research activities of the Institute’s programmes. The papers are internally refereed 
and are disseminated to inform and invoke debate on policy issues. Opinions 
expressed in the papers are entirely those of the authors and do not necessarily 
reflect the views of the Institute.
This paper is produced under the KIPPRA Young Professionals (YPs) programme. 
The programme targets young scholars from the public and private sector, who 
undertake an intensive one-year course on public policy research and analysis, and 
during which they write a research paper on a selected public policy issue, with 
supervision from senior researchers at the Institute.
ii
Abstract
Kenya ushered a devolved system of governance in 2013 leading to the 
formulation of two levels of governance, one national government and 47 county 
governments 27 of which are classified as ASALs. A total of 14 functions were 
devolved inclusive of healthcare. This study assessed the effect of devolution 
on the healthcare service delivery in the ASAL counties. This was achieved 
through a comparison of selected indicators of healthcare access before and 
after devolution. Wilcoxon signed-rank test is also applied to the data to help 
understand whether a significant change was experienced after devolution. 
Results indicate a significant increase in the number of health professionals, 
number of beds, and number of hospitals. However, individuals covered longer 
distances, drug unavailability in public health facilities increased, and there 
were higher incidences of catastrophic health expenditure. The number of 
patients satisfied with the privacy in outpatient and inpatient care also reduced 
significantly after devolution. Counties are encouraged to pass the facility 
improvement fund bill and embrace the digitalization of healthcare as a means 
of raising funds and reducing the distance to the health facilities. In addition, it 
is important to educate healthcare workers on the importance of maintaining 
patient confidentiality to improve acceptability. Also educate the masses on the 
importance of preventive healthcare to help curb the burden of catastrophic 
health expenditure.
iii
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Abbreviations and Acronyms
ASALs         Arid and Semi-Arid Lands
HIV            Human Immunodeficiency Virus
IP              Inpatient
KHHEUS     Kenya Household and Health Expenditure Utilization Survey
Kms           Kilometres
Ksh            Kenya Shilling
MoH           Ministry of Health
NGO           Non-Governmental Organizations
OP            Outpatient
SSA           Sub-Saharan African
UHC           Universal Health Coverage
WHO         World Health Organization
iv
Table of Contents
Abstract ....................................................................................................................ii
Abbreviations and Acronyms .................................................................................iii
List of Tables ........................................................................................................... iv
List of Figures ..........................................................................................................v
1 Introduction .......................................................................................................... 1
3. Literature Review ................................................................................................5
3.1 Theoretical Literature- Conceptual Framework ...............................................5
3.2 Empirical Literature ..........................................................................................6
4.  Methodology .......................................................................................................7
1.1 Accessibility, Availability, Acceptability and Quality (AAAQ) of Healthcare ... 7
1.2 Wilcoxon-signed rank test .................................................................................8
2 Results and Discussions ..................................................................................... 12
3 Conclusion and Policy Recommendations .........................................................23
3.1 Conclusion  .......................................................................................................23
3.2 Policy Recommendations ................................................................................23
References ..............................................................................................................25
v
List of Tables
Table 3.1 Description of variables……………………………………………………….….......…12
Table 4.1 Accessibility ............................................................................................ 13
Table 4.2: Availability ............................................................................................ 16
Table 4.3: Acceptability ......................................................................................... 19
Table 4.4:  Quality .................................................................................................20
List of Figures
Figure 2.1: Levels of healthcare facility ...................................................................4
Figure 4.1: Choice of OP and IP care by type of facility ........................................ 14
Figure 4.2: Reasons for bypassing a health facility .............................................. 14
Figure 4.3: Distribution of beds and facilities by county in 2013 .........................17
Figure 4.4: Distribution of beds and facilities in ASALs by county in 2018 ........ 18
Figure 4.5: Average number of equipment per 10,000,000 population in 2013 and 
2018 ....................................................................................................................... 21
1. Introduction
The Constitution of Kenya created a devolved system of government which saw 
power, resources, and representation moved to the local level. Article 174(c) and 
(f) of the Constitution of Kenya outlines that among the objectives of devolution 
is to enhance public participation of the people in the issues affecting them and to 
make services easily accessible throughout the country respectively. When citizens 
are involved in the decision-making process, they are deemed to have achieved the 
highest level of public participation.
The 14 functions were devolved namely, agriculture, county health services, 
control of pollution, cultural activities, county transportation, animal control, 
trade development, county planning and development, pre-primary education 
and polytechnics, implementation of specific national policies, county public 
works, firefighting services, control of drugs and pornography and coordinating 
the participation of communities.
Of these 14 devolved functions this study will focus on assessing how devolution 
has impacted the health sector since it is one of the functions that was greatly 
devolved to the counties including human resource and staffing, planning, 
budgeting, acquisition of drugs and medical supplies to county referral hospitals, 
health centres, dispensaries, and community units. The national government 
was only mandated with the control of the national referral hospitals and the 
formation of policies governing the health sector. The healthcare sector is also 
important because improvement in health has been associated with an increase 
in the productivity level of individuals and the economic growth of economies 
(World Bank Group, 2022).
Before devolution, Kenya was under a centralized system of governance 
characterized by unbalanced regional resource allocation, limited involvement of 
the people in the governance process, and marginalization of some communities, 
especially ASALs Wanyande & Mboya, (2016). The devolved system of governance 
is thus expected to aid in the improvement of service delivery, improvement in the 
public participation process, and a reduction in marginalization among the various 
regions of the country Pietrzyk et al. (2018). In Kenya, the devolved system of 
governance was adopted in 2013 after the promulgation of the 2010 constitution. 
Data from the 2018 Kenya Household Health Expenditure Utilization survey 
(KHHEUS) indicates that 19 per cent of the people travelled over 10 kilometres 
to seek health services, a 6 per cent increase from 2013. In addition, the Out-of-
Pocket expenditure in the healthcare sector increased by 90 per cent indicating 
that devolving healthcare to the counties had not made it more affordable. Four 
years into devolved governance in Kenya, Khaunya et al. (2015) found that 
counties were still not experiencing notable improvement in service delivery at 
the county level as they were marred with challenges of corruption, non-payment 
of staff salaries, and duplication of roles.
It is against this backdrop that this study seeks to analyze the effect of devolution 
for the period 2013-2023 on the service delivery of ASALs in Kenya specifically the 
health sector. The study seeks to consider what areas of devolution have worked 
1
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
well, and what areas could be improved by comparing the current performance 
against the set international standards and targets.
This study aims to analyze the progress that has been made in the healthcare sector 
and challenges encountered in Kenya since it was devolved. Specific objectives 
include:
(i) To assess the contribution of devolution to service delivery in the healthcare 
sector of the ASAL counties.
(ii) To identify areas of improvement in county healthcare delivery by analysis 
of the existing gaps based on the recommendations of WHO and MOH.
The rest of the paper is organized as follows; Section 2 discusses the stylized facts 
and literature review, methodology is discussed in section 3, Section 4 discusses 
empirical findings and results and finally Section 5 outlines conclusions and policy 
recommendations.
2
2.  The Journey of Devolution of Healthcare Function
Kenya’s healthcare sector has been evolving towards decentralization since the 
1960s, long before the 2010 constitution formalized the process. Early steps 
included the formation of District Health Boards and decentralized units within 
the Ministry of Health. These efforts laid the groundwork for a significant 
transformation under the new constitution, aiming to overcome previous 
challenges such as access disparities, resource limitations, and inefficiencies. The 
goal is a more accessible, equitable, and community-driven healthcare system in 
Kenya.
The health sector was highly devolved, and the responsibilities shifted to the 
county governments included budgeting, planning, and financial management, 
human resource management, provision of emergency medicine and medical 
supplies, service delivery focusing on disease prevention and health promotion, 
management and operation of county hospitals, dispensaries, and health centres, 
and ambulance facilities. 
Figure 2.1: Levels of health care facility
3
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
The national government was left in charge of policy creation, quality assurance, 
and the management of the national referral hospitals including Kenyatta National 
Hospital, Moi Teaching and Referral Hospital, National Spinal Injury Hospital, 
Mathari National Teaching and Referral Hospital, and Kenyatta University 
Teaching and referral hospital. The hospitals being managed by the counties are 
divided into three tiers with the county referral hospitals forming the third tier, the 
second tier is composed of health centres, dispensaries, and nursing homes, while 
the first tier is made up of community units. Medicine procurement in county-
managed hospitals is achieved through the county pharmacist who consolidates 
all the medicine requirements from all health facilities and submits it to the county 
treasury for purchase. Before devolution, the district health management team 
oversaw these purchases. Although the adoption of obtaining medicine through 
the county treasury was aimed at making the process more efficient, it ended up 
slowing the process and many counties soon ran out of stock RESSYT (2016).
Another significant change that occurred in the healthcare system was that 
hospitals no longer received funds directly from the national government and had 
to rely on user fees and county government accounts. These changes among many 
others affected the service delivery in the county health system. The ability of the 
hospitals to respond to emergencies, provide quality medicine and services, procure 
essentials, and motivate staff was affected. The tier-one healthcare facilities were 
even more hard hit because, at the onset of devolution, the president announced 
free maternity services in these facilities greatly reducing their user fees.
The national government introduced conditional grants to the county healthcare 
facilities to help in funding them. Examples of conditional allocations in health 
disbursed through Treasury is the level 5 grant which sought to assist counties 
with level 5 hospitals and User Fees foregone conditional grant. The Equalization 
Fund is an example of conditional grant which was established by the constitution. 
At the onset of devolution, 14 counties were identified as marginalized including 
Turkana, Mandera, Wajir, Marsabit, Samburu, West Pokot, Tana River, Narok, 
Kwale, Garissa, Kilifi, Taita Taveta, Isiolo, and Lamu. To fast track the integration 
of these counties into the economic and social activities of the county, equalization 
fund was established in chapter 12 of the constitution. This stipulates that 0.5 per 
cent of all revenue collected by the national government be used to provide basic 
amenities like roads, health facilities and electricity to marginalized areas. 
In the first three years of devolution 2011/2012 to 2013/2014, no allocations were 
made since there were delays in allocation of equalization fund with non-existence 
of a criteria defining how the fund should be shared among the marginalized 
areas (Government of Kenya, 2022). Between 2015 and 2017, Ksh 10.3 billion was 
allocated to 14 counties that had been identified as marginalized. In the second 
policy on marginalization, however, 20 more counties were included leading to 
a total of 34 counties. Of the projects implemented in the first disbursement, 23 
per cent were in the health sector, 23 per cent in infrastructure, 7 per cent were in 
irrigation, 36.3 per cent in water and sanitation, 5 per cent in energy, 2.8 per cent 
in early learning and the remainder in vocational training across the marginalized 
counties. The completion rate for the health projects was at 10 per cent with 
counties citing inadequate funding as the reason for non-completion. 
4
3. Literature Review
This section focuses on the status of devolved healthcare and an empirical 
literature review of devolved healthcare at local and international level.
3.1 Theoretical Literature- Conceptual Framework
Decentralization theorem 
This theorem was brought forward by Wallace Oates in 1972. The theorem states 
that provision of local public goods and service delivery could be made more efficient 
when left to lower levels of government. He argues that the local government has a 
more superior understanding of the preferences and wants of its members and so 
the central government is unable to match policies and goods to the regions where 
they are needed most. Decentralization is thus argued as one way of matching public 
goods and services to the local needs while reducing costs. The factors put forward 
in this study when deciding which functions to decentralize include economies of 
scale in the production of public goods, the potential of interregional transfer of 
goods, and heterogeneity of households across the different regions. Devolution 
enables healthcare services to be more responsive to local health issues, and social 
determinants of health. Local governments can tailor healthcare services to the 
specific needs and demographics of their populations. This flexibility allows for 
more responsive and targeted healthcare interventions This responsiveness can 
lead to more effective public health interventions.
Principal-Agent theory
According to this approach, the principal has defined objectives to be met. The 
agents are the means through which these objectives are achieved. However, the 
agents also have their self-interests and may have more information compared 
to the principals. They may thus end up pursuing their interests secondary 
to those of the principal. The principal is aware of the limitations of the agent 
and may seek to obtain more information from the agent, but this comes with 
higher cost implications. The principal thus may resort to alternative means of 
gaining information from the agent including close monitoring and punishment. 
In devolution of health. The Ministry of Health which is mandated on behalf of 
the citizens, is the principal whose objective is promoting equity, efficiency, and 
affordability in the healthcare sector whereas the localized health facilities in 
the counties are the agents through which the principal meets these objectives 
(Bossert ,1998).
3.2 Empirical Literature 
The effect of devolution on healthcare access is not a one-size-fits-all all but 
rather differs across countries and counties within the same country. According 
to Abimbola et al., (2019), this depends on various institutional, socioeconomic, 
and geographical factors like the availability of quality health infrastructure 
5
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
and services, the level of economic development of a region, and the degree of 
autonomy of the local governments. They outline that devolution is likely to 
improve healthcare access if individuals can choose the facility that best suits their 
needs, when local governments tailor health interventions to the needs of their 
population, and when a strong civil society exists to hold the local government 
accountable for their actions. 
Wanzala & Oloo (2019) in a study on the role of devolution in healthcare service 
delivery and health workforce in Kenya, opine that devolution may improve 
citizen participation, monitoring, and delegation of roles in the healthcare sector, 
and accountability. However, human resource deficiency and interference from 
the central government are identified as some of the elements that currently slow 
down the healthcare sector in Kenya. In the study, ways of increasing the effect 
of devolution like ensuring transparency and accountability and creating ethical 
standards to be followed are identified. Hémet et al., (2023) also examine how 
healthcare services in Kenya changed after devolution and find that the use of 
public health facilities especially clinics, increased after devolution. However, 
they outline the strong role played by homogeneity in the counties as services 
became more affordable and accessible for the dominant ethnic group in the 
county. This study also acknowledges the role played by devolution in making local 
governments more responsive to the needs of the people. Similarly, Makokha & 
Amis (2017) also found that devolution improved healthcare delivery in Makueni 
and Kisumu counties of Kenya. The channels identified in this study include 
reduced bureaucracy in the procurement of healthcare facilities and bringing 
services closer to the people.
Although different countries usually have different levels and types of 
decentralization and devolution, McCollum et al., (2018)conducted a comparative 
study on how devolution affected ten selected counties in Kenya and one district 
in Indonesia. Key informants’ interviews and focus group discussions methods 
used in obtaining the data. Similarities that arose in the two countries regarding 
the effect of devolution include a lack of clear guidance on how to implement 
devolution, limited capacity of decision-makers, and prioritization of curative 
health services in place of preventive measures. Differences were identified in 
the healthcare structure of these two countries in that whereas in Kenya only the 
policy formulation was left at the national level, in Indonesia, policy formulation 
and regulation were left at the central and provincial governments. 
Differently, Kairu et al., (2021)  examined how health facilities in Kenya were 
financed after devolution. The results revealed that planning and budgeting 
were not standardized across counties, and that led to public hospitals and 
health centres relying on donor funds and user fees for running the facilities. 
Recurrent expenditures more so staff salaries were found to take a larger portion 
of the expenditure, and this limited the amounts available for infrastructure 
development and purchase of drugs and equipment.
6
4.  Methodology
This section presents the estimation methodology and describes the data used. 
AAAQ Framework
Health service delivery is analysed based on the AAAQ framework set by the 
WHO on assessing the right to health. These include Accessibility, Availability, 
Acceptability, and Quality of healthcare in the ASAL counties. This framework is 
chosen because it is commonly used in describing health care delivery.
Accessibility
Accessibility of health services is determined by the ability of a health system 
to ensure health services are available to all. Accessibility of healthcare can be 
categorized in terms of physical and economic access. 
Physical accessibility means that the healthcare facilities are within a reachable 
distance which has been set by the WHO at a 5km radius. When healthcare 
facilities are physically accessible, health service delivery is likely to increase. For 
this study, physical accessibility of healthcare was analyzed based on the distance 
covered by a patient in reaching both inpatient and outpatient health facilities in 
2013 and 2018 categorized by ownership since this determines the speed at which 
one reaches a health institution during medical emergencies. The mean distance 
from a health facility per county is calculated and comparisons made to determine 
whether they have increased or decreased.
Economic accessibility involves an analysis of catastrophic health expenditure per 
county. Healthcare expenses are defined as catastrophic when the amount spent 
on health exceeds 40 per cent of the total amount spent on non-food items or 10 
per cent of the total expenditure (Maina, 2015) . The study calculates catastrophic 
health expenditure per county based on the 40 per cent expenditure on non-
food items threshold. Comparisons are then made for the period before and after 
devolution to determine whether catastrophic health expenditure has increased 
or not.
Availability
The availability of healthcare requires that there are enough healthcare facilities 
and infrastructure for everyone, and they have competencies to match the needs 
of the population. In the study, the availability of healthcare before and after 
devolution was indicated by the number of health facilities per 100,000 people 
per county. The number of beds available in the health facilities before and after 
devolution is also analysed as it could be indicative of the availability of basic 
infrastructure in the health facilities. The significance of the change in before 
and after values is then determined using a rank test. In addition, comparisons 
7
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
will be made with the WHO recommendations to determine how the counties are 
performing.
Acceptability
Acceptability of healthcare refers to the extent to which the healthcare provided 
takes into consideration, privacy, and confidentiality of the patient to preserve 
human dignity and complies with the existing medical ethics and socio-cultural 
norms in place. When patients are treated in privacy and dignity, they are likely 
to develop trust in the healthcare system and so open to the providers regarding 
their health issues. On the other hand, patients may fail to utilize the services of 
a health facility that is accessible and available if they feel that their privacy is 
violated. This implies that privacy is an important aspect of acceptability of health 
service. In this study, acceptability of healthcare before and after devolution was 
indicated by the percentage of patients who were satisfied with the outpatient 
and inpatient care received from both private and public health facilities. The 
significance of the change in acceptability is the determined using rank test.
Quality
Quality of healthcare involves an assessment of health worker competencies, skills 
and knowledge which influences the nature of treatment received by the patients 
from the healthcare workers, and whether it meets the required standards to meet 
the desired health outcomes. This involves an assessment of the trained personnel 
in the health facilities, availability of drugs, equipment and infrastructure 
including MRI, ultrasound, x-ray machines, and ICU beds per county. In the 
study, the quality of healthcare was analyzed based on the number of trained 
healthcare personnel per county for the period before and after devolution, 
percentage of patients who found the drugs prescribed to them at the facility, and 
the average number of equipment available per county per 10,000 population. 
An increase in the number of skilled healthcare providers, number of healthcare 
equipment per 10,000 population, and percentage of patients who find all the 
medicine prescribed for them could be an indicator of improvement in the quality 
of healthcare in the counties after devolution. 
Wilcoxon-signed rank test
A quantitative approach involving a nonparametric method of analysis, the 
Wilcoxon signed-rank test was adopted. Nonparametric approaches can be used 
to analyze data with large variances as they are not constrained by the distribution 
of the population Scheff (2016). The Wilcoxon-signed rank test is designed for the 
comparison of two paired samples taken from a population that does not assume 
normality or near normality in distribution Martín et al., (2016) & Rosner et al., 
(2006). It is useful in measuring the change in score between two paired samples 
at different times. This test is preferred because it is parameter free implying 
8
Methodology
that the underlying assumptions are always met. The samples are related and 
obtained from the same population for two years 2013 and 2018 indicative of the 
period before and after devolution respectively. In this study, it was applied to test 
whether a statistically significant change occurred after devolution.
The Wilcoxon signed-rank test compares two related samples and tests for any 
difference in their population means ranks Rey & Neuhäuser, (2011) & Scheff 
(2016). The test is appropriate for repeated tests where the same observation is 
analyzed under two different conditions in this case before devolution and after 
devolution.
Let there be two paired samples x1, x2, x3,………………., xn and y1, y2, y3,……..yn of a 
sample of size n from a continuous population with probability density function 
F, and median M.
Then Di=xi-yi, i=1,2,3,…………,n which represents the difference between two 
paired random variables. Where Di is assumed to be mutually independent, comes 
from a continuous population with probability function F, and is symmetric about 
the median M.
The Wilcoxon signed-rank test statistic is
W=∑_(i=1)nRi .signDi
Where Di= Xi- Yi , Ri is the rank of |Di |, i=1,2,3,…………,N and sign(Di) =1 if Di>0 
and sign(Di) =-1 if Di<0
The differences are obtained in their absolute values and assigned ranks 
from the smallest to the largest. The ranks corresponding to the positive and 
negative differences are then summed up separately. The total of the rank 
values corresponding to the positive and negative differences are R_+ andR_- 
respectively. A difference is ascertained to exist between the two samples if R_+ or 
R_- values are lower than the critical value w(z) calculated as follows:
W(z)= (n(n+1))/4 -√(z&(n(n+1)(n+2))/24)    where z is tabulated based on the 
level of significance desired. The desired number of n should be greater than 10. 
The null hypothesis is that H0:M=0 (The distribution of the differences Di is 
symmetric about 0.
The alternative hypothesis is that H1 M≠ 0 
It tests the hypothesis that the median of the various health indicators for 
availability, quality and accessibility is zero for the period before and after 
devolution. The outcomes of interest are the number of hospitals per county, the 
number of beds in the hospitals, the number of qualified health personnel in the 
counties, the prevalence of malaria, the distance covered to reach a health facility 
in kilometres, and the time taken to reach a health facility in minutes.
The Wilcoxon signed rank test is implemented in this study in the following steps:
First, the desired observations with values in 2013 and 2018 indicative of the 
period before and after devolution are presented. Second, the absolute differences 
between the values for the two periods are obtained and ranked then summed 
9
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
up to obtain the values of R_+ and R_-. Third, a statistical value of z is obtained 
for the study which is taken to be 1.96 corresponding to the 95 per cent level of 
significance. Finally, a value for w(z) is obtained and compared with the R_+ and 
the R_- values. This is repeated for each indicator using the values for before and 
after devolution.
Data and Variables
Data for the empirical analysis was collected from Kenya National Bureau of 
Statistics (KNBS) issues of cross-sectional survey data, KHHEUS 2013 and 
2018, and census 2009 and 2019. The data was for two periods 2013 and 2018 
representing the period before and after devolution respectively. Although the 
KHHEUS data was collected for individuals, the data is aggregated into counties 
before being analyzed. 
26 ASAL and 18 non-ASAL counties were included in the study. Three ASAL 
counties, Mandera, Garissa, and Wajir are not included in the analysis as they 
were not included in the 2013 KHHEUS data survey. The use of more than one 
dataset in the analysis was based on the fact that there was no single dataset 
which had all the indicators for the two time periods. The density of beds and 
healthcare workers per 10,000 population was not disaggregated by ownership of 
the facilities per county because of missing data in some counties for the period 
under consideration.
Table 3.1: Description of variables 
Measure of service Variable Name Definition and Source
delivery measurement
Accessibility Distance The average KHHEUS 2013 
distance covered and 2018 survey
when seeking 
OP health care 
per county by 
facility ownership 
measured in Kms.
The average KHHEUS 2013 
distance covered and 2018 survey
when seeking 
IP health care 
per county by 
facility ownership 
measured in Kms.
10
Methodology
Cost Average KHHEUS 2013 
catastrophic and 2018 survey
health 
expenditure per 
county by facility 
location measured 
by percentage of 
expenditure on 
health compared 
to non-food items.
Availability Hospitals Facility density Health sector 
per county annual 
per 10,000 performance 
population by review report 
ownership. 2020/21
Beds Number of beds Health sector 
per hospital annual 
measured performance 
per 10,000 review report 
population. 2020/21
Acceptability Privacy Percentage KHHEUS 2013 
of patients and 2018 survey
per county by 
ownership who 
expressed their 
satisfaction with 
the privacy of OP 
services received.
Percentage Health sector 
of patients annual 
per county by performance 
ownership who review report 
expressed their 2020/21
satisfaction with 
the privacy of IP 
services received.
Quality Workers Number of Health sector 
trained health annual 
workers in performance 
hospitals per review report 
100,000 people. 2020/21
11
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Drug availability Percentage of KHHEUS 2013 
patients who and 2018
found all of the 
drugs prescribed 
to them in a 
health facility by 
ownership per 
county.
Equipment Number of MRI Health sector 
availability machines per annual 
county per 10,000 performance 
population review report 
2020/21
Number of CT Health sector 
scans per county annual 
per 10,000 performance 
population review report 
2020/21
Number of Xray Health sector 
machines per annual 
county per 10,000 performance 
population review report 
2020/21
Number of Health sector 
Ultrasound annual 
machines per performance 
county per 10,000 review report 
population 2020/21
Number of ICU Health sector 
beds per county annual 
per 10,000 performance 
population review report 
2020/21
12
4. Results and Discussions
In this section, the results of the mean differences per county for both ASALS 
and non-ASALS and Wilcoxon signed rank test are presented. The results are 
presented separately for the different measures of service delivery including 
accessibility, availability, acceptability, and quality.
The results indicate that on average, people travelled longer distances to seek OP 
and IP care from private health facilities compared to public in both ASAL and 
non-ASAL counties. Longer distances were also covered to reach public health 
facilities after devolution. The distance covered to seek IP services from a private 
health facility however, reduced in both ASALs and non-ASALs. The average 
distance covered in counties, however, is higher than the 5km radius that has been 
recommended by WHO. The average catastrophic health expenditure is higher 
after devolution compared to before devolution in both rural and urban areas.
The Wilcoxon signed rank test results that test for the significance of the changes 
to accessibility after devolution indicate that the increase in the average distance 
covered by patients seeking OP and IP care from public health facilities after 
devolution was found to be significant in ASAL counties. Although the distance 
covered to both public and private facilities also increased in the non-ASL areas, 
the increase was not as significant as compared to ASALs. The incidence of 
catastrophic health expenditure increased significantly across the counties after 
devolution in both urban and rural areas. However, incidences of catastrophic 
health expenditure were higher in ASALs compared to non-ASALs. Results indicate 
that both physical and economic accessibility have not improved significantly 
in the first five years as individuals cover longer distances and spend a higher 
percentage of their income on health compared to non-food items when seeking 
IP and OP healthcare services.
The increase in distance covered to seek OP and IP care after devolution, however, 
is attributable to the fact that people seek healthcare from their preferred facility 
which may not necessarily be the nearest facility to them. The preference is 
influenced by other factors including availability of drugs, health infrastructure 
and equipment, referrals, and qualified medical personnel. The increase in the 
incidences of catastrophic health expenditure which was more prevalent in the 
urban areas compared to rural calls for the need to train individuals on the need 
for preventive healthcare as opposed to curative health. The national government 
in a bid to curb these incidences of catastrophic health expenditure in the counties, 
rolled out the universal healthcare coverage (UHC) pilot in four counties, Isiolo, 
Kisumu, Nyeri, and Machakos. Conditional grants were given to level 4 and 5 
facilities in these counties while households receive treatment at no cost. The 
UHC policy, however, includes strengthening mandatory pre-paid health sources 
like NHIF for all households in the country. 
13
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
14
Table 4.1: Accessibility
ASALs Non-ASALs
Variable 2013 Mean 2018 Mean Wilcoxon Z No. of obs 2013 Mean 2018 Wilcoxon No. of obs
(Std Dev) (Std Dev) statistic (Std Dev) Mean Z statistic
(Std 
Dev)
Distance to public OP health facility in Kms 7.91 11.26 -2.65** 26 5.88 6.52 -0.61 18
(3.39) (6.59) (1.66) (2.68)
Distance to private OP facility in Kms 11.69 16.08 -2.68** 26 9.49 9.78 -0.19 18
(5.28) (8.12) (4.50) (5.46)
Distance to public IP health facility in kms 14.78 33.30 -3.67*** 26 17.71 18 -0.45 18
(8.77) (20.63) (11.91) (10.17)
Distance to private IP health facility in Kms 41.09 35.5 -0.87 26 51.58 19.83 2.63** 18
(63.72) (26.86) (48.62) (11.84)
Catastrophic health expenditure in urban areas 0.77 4.99 -3.97*** 26 1.46 1.73 -0.91 18
(0.53) (13.23) (1.15) (1.37)
Catastrophic health expenditure in rural areas 1.42 3.92 -2.94*** 26 1.44 2.11 -0.65 18
(0.69) (6.52) (1.30) (2.63)
***, **, * significance at 1, 5, and 10 % respectively
Source: Authors computation using Stata
Results and discussions
Figure 4.1: Choice of OP and IP care by type of facility
Source of data: KHHEUS survey 2013 and 2018
Figure 4.1 reveals that public health facilities including county government 
hospitals, government dispensaries, and government health centres were the 
most preferred by those seeking OP care across all counties followed by private 
hospitals. For those seeking IP care, the preferred health facilities were the 
county government hospitals followed by private and faith-based hospitals whose 
preference varies across the counties. This implies that the preference for public 
health facilities, especially the county government hospitals is still high for those 
seeking IP and OP care across the counties.
15
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Figure 4.2: Reasons for bypassing a health facility
Source of data: KHHEUS survey data 2013 and 2018
The main reason for bypassing a nearby health facility in 2013 as revealed in 
the study is the absence of specialists at the facilities which stood out in both 
private and public facilities. However, this is not the case in 2018 when medicine 
unavailability is seen as the main reason for bypassing a health facility. This could 
be an indication that more specialists and qualified personnel were employed in 
both private and public health facilities. The higher percentage of people who 
bypassed a health facility due to medicine unavailability indicates the key role it 
plays in healthcare access. Being referred to another health facility was also one of 
the main reasons stated for not getting healthcare access in a nearby facility and 
this is based largely on the availability of infrastructure and equipment to handle 
the health issue at hand.
16
Results and discussions
17
Table 4.2: Availability 
ASALs Non-ASALs
Variable 2013 Mean 2018 Mean Wilcoxon No. of 2013 2018 Wilcoxon Z No. of obs
(Std Dev) (Std Dev) Z statistic obs Mean Mean statistic
(Std Dev) (Std Dev)
No. of beds per 10,000 population 18.69 20.95 -2.065** 29 26 32.27 -3.298** 18
(10.41) (14.19) (29.49) (31.75)
Density of total facilities per 10,000 2.25 2.16 0.825 26 1.63 1.72 -1.32 18
(0.66) (0.61) (0.47) (0.48)
Density of public facilities per 10,000 1.32 1.23 1.61 26 0.87 0.92 -1.80 18
(0.51) (0.45) (0.32) (0.28)
Density of private facilities per 10,000 0.93 0.92 -0.48 26 0.76 0.80 -1.41 18
(0.47) (0.49) (0.56) (0.50)
***, **, * significance at 1, 5, and 10 % respectively
Source: Authors computation using Stata
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Results indicate that although the average number of beds per 10,000 population 
per county has increased in both ASAL and non-ASAL areas, the increase is greater 
in non-ASALs as compared to ASALs for the periods before and after devolution. 
The overall facility density indicates an increase in the number of facilities per 
10,000 population in the non-ASAL regions but a slight decrease in the density 
of facilities in the ASAL regions. Based on ownership, the data reveals that more 
private and public facilities were built in the non-ASAL areas compared to ASALs 
after devolution. 
The Wilcoxon signed rank test results that test for the significance of the changes 
to the availability of healthcare after devolution indicate that the number of beds 
per 10,000 population increased significantly in ASAL and non-ASAL counties. 
However, the overall increase in the number of facilities after devolution was not 
significant in both ASALs and non-ASAL counties. However, the density of public 
facilities is higher than that of private facilities in both ASALs and non-ASALs. 
An in-depth examination of the total bed count and healthcare facilities in the 
ASAL counties provides insight into the measures that have been implemented to 
enhance healthcare accessibility. The distribution of beds and health facilities per 
10,000 population for each county is depicted in Figures 4.3 and 4.4 for the years 
2013 and 2018, respectively. The World Health Organization's recommended 
bed and facility density of 25 and 2.2 per 10,000 population, respectively, are 
represented by red lines.
The counties are categorized into four quadrants. The bottom left quadrant 
represents counties that have not met the recommended bed and facility count. 
The bottom right quadrant includes counties that have achieved the recommended 
facility count but not the bed count. Counties that have met both the recommended 
bed and facility count are in the top right quadrant, while those that have met the 
bed count but not the facility density per 10,000 population are in the top left 
quadrant.
This quadrant analysis reveals a decline in healthcare availability in Elgeyo 
Marakwet, Marsabit, Isiolo, Kwale, and Kilifi counties post-devolution, as 
evidenced by their shift from the bottom right quadrant in 2013 to the bottom 
left quadrant in 2018. Conversely, Embu and Makueni counties demonstrated 
significant improvements in healthcare availability, as indicated by their 
transition from the bottom right to top right quadrant and bottom left to bottom 
right quadrant, respectively.
Further, a slight improvement is observed in the number of counties meeting 
the recommended number of beds and facilities rising from three in 2013 to four 
in 2018. However, it's important to note that out of the 14 counties identified as 
marginalized, eight (including Turkana, Marsabit, West Pokot, Tana River, Narok, 
Kwale, Kilifi, and Isiolo) all fall in bottom left quadrant in 2018 falling short of 
both the recommended number of beds and facilities per 10,000 population. 
This suggests that despite the differences among counties, common challenges 
such as marginalization can hinder healthcare service delivery. In addition, not 
all counties started at the same level when devolution began. Five years into 
devolution, marginalized counties continue to trail behind non-marginalized ones.
18
Results and discussions
The results also indicate that only four counties, Meru, Machakos, Nyeri, and 
Embu have achieved the recommended number of beds and facilities while 
marginalized counties like Turkana, Isiolo, Marsabit, West Pokot, Tana River, 
Narok, and Kwale have less than the recommended number of beds and facilities. 
Increased health service availability is observed in Embu and Makueni counties 
while decreased availability is observed in Elgeyo Marakwet, Marsabit, Isiolo, 
Kwale, and Kilifi counties. The availability of the recommended number of beds 
and health facilities services in Meru, Machakos, Nyeri, and Embu counties can 
be attributed to the presence of level 5 facilities in these counties which qualify 
them to receive the level 5 conditional grant from the national government since 
this is an additional fund to these counties. The improvement in Makueni county 
is attributable to the presence of Makueni care where households contribute Ksh 
500 annually and the county retains these funds for providing services to the 
members and improvement of at the healthcare facilities. Makueni county have 
also implemented the facility improvement fund which allows public hospitals to 
retain and use the user collected funds. The reduction in availability in Elgeyo 
Marakwet, Marsabit, Isiolo, Kwale, and Kilifi counties could be attributed to 
unavailability of level 5 facilities in these counties implying fewer grants to 
improve the healthcare sector. Further, Kilifi county had not implemented the 
facility improvement fund in 2018 in spite of enacting it. Isiolo county on the other 
hand, had not yet implemented the facility improvement fund and channelled 
their funds to the county revenue fund for redistribution (Mbuthia et al., 2019).
Figure 4.3: Distribution of beds and facilities by county in 2013
Source of data: Health sector annual performance review report 2020/21
19
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Figure 4.4: Distribution of beds and facilities in ASALs by county in 
2018
Source of data: Health sector annual performance review report 2020/21
Results indicate that the percentage of people who expressed satisfaction with 
20
Results and discussions
21
Table 4.3: Acceptability 
ASALs Non-ASALs
Variable 2013 Mean 2018 Mean Wilcoxon Z No. of obs 2013 Mean 2018 Mean Wilcoxon Z No. of obs
(Std Dev) (Std Dev) statistic (Std Dev) (Std Dev) statistic
% of people who were satisfied with the 79.15 71.16 2.62** 26 77.91 66.99 3.41** 18
privacy in OP care in public facilities
(7.15) (9.21) (6.63) (7.60)
% of people satisfied with privacy in OP 55.20 37.53 4.15*** 26 52.18 37.11 3.201** 18
care in private facilities
(15.64) (11.73) (14.09) (10.53)
% of people satisfied with privacy in 71.95 67.69 1.25 26 69.99 60.74 2.86** 18
public IP care
(9.50) (14.37) (7.87) (11.24)
% of people satisfied with privacy in 53.39 49.11 1.46 26 53.92 54.62 0.15 18
private IP care
(13.82) (15.27) (15.76) (10.37)
***, **, * significance at 1, 5, and 10 % respectively
Source: Authors computation using Stata
the privacy in the OP and IP care from various public and private health facilities 
areas was higher before devolution compared to after devolution for both ASALs 
and non-ASALs. The mean satisfaction in the private health facilities was lower 
than the mean satisfaction in the privacy received when obtaining care in public 
facilities for both ASALs and non-ASALs.
The Wilcoxon signed rank test results that test for the significance of the changes 
to the acceptability of healthcare after devolution indicates the existence of a 
significant decrease in the percentage of people who were satisfied with privacy 
received while seeking OP care in public and private facilities for both ASALs and 
non-ASALs. The percentage decrease in the number of people seeking IP care 
who were satisfied with the privacy received in public and private facilities was 
insignificant. 
A report by the Commission for Constitution Implementation (2015)  indicated 
that privacy in some outpatient facilities in Kitui County was lacking especially 
for the case of HIV patients who had designated treatment areas and the choice of 
placement of equipment like condom dispensers in the reception as opposed to the 
washrooms. This trend when replicated in other counties could discourage people 
from accessing and utilizing healthcare from these facilities thus exacerbating the 
percentage of people who were satisfied with the privacy received while seeking 
healthcare.
Results and discussions
23
Table 4.4: Quality
ASALs Non-ASALs
Variable 2013 Mean 2018 Mean Wilcoxon Z No. of 2013 Mean 2018 Mean Wilcoxon Z No. of obs
(Std Dev) (Std Dev) statistic obs (Std Dev) (Std Dev) statistic
Number of MRI machines per 0.003 0.008 -0.89 26 0.008 0.007 0.24 18
10,000 population
(0.005) (0.017) (0.01) (0.008)
Number of CT scan machines per 0.012 0.027 -2.84*** 26 0.013 0.035 -2.72*** 18
10,000 population
(0.019) (0.027) (0.013) (0.029)
Number of Xray machines per 0.007 0.059 -4.18*** 26 0.009 0.045 -3.72*** 18
10,000 population
(0.01) (0.044) (0.011) (0.018)
Number of Ultrasound machines 0.061 0.101 -3.11*** 26 0.05 0.097 -2.67** 18
per 10,000 population
(0.044) (0.063) (0.036) (0.025)
Number of ICU beds per 10,000 0.012 0.107 -3.52*** 26 0.029 0.151 -3.13*** 18
population
(0.019) (0.169) (0.06) (0.163)
% of people seeking OP care in 87.13 81.51 3.31*** 26 83.97 80.04 2.59** 18
public facilities who found the drugs 
(5.26) (6.73) (6.62) (7.28)
needed
% of people seeking OP care in 96.85 96.42 0.62 26 96.85 96.42 0.54 18
private facilities who found the 
(1.78) (2.09) (1.78) (2.09)
drugs needed
Number of health workers per 13.01 17.30 -4.314*** 29 13.22 17.03 -3.158** 18
10,000
(7.57) (7.81) (8.04) (6.70)
***, **, * significance at 1, 5, and 10 % respectively
Source: Authors computation using Stata
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Results indicate that on average, there were 3 MRI machines for every 10,000,000 
people in the ASAL counties compared to 8 for the non-ASALs in 2013, this 
number increased to 8 in the ASAL counties and 7 in the non-ASAL counties in 
2018. The average number of CT scans per county increased from 12 to 27 per 
10,000,000 in 2018 in ASALs and from 13 to 35 per 10,000,000 in non-ASALs. 
X-ray machines increased from 7 to 59 per 10,000,000 in 2018 in the ASALs and 
from 9 to 45 in the non-ASAL counties. Similarly, ultrasound machines increased 
from 61 to 101 per 10,000,000 population in 2018 in ASALs and from 5 to 97 
in non-ASAL counties. The number of ICU beds increased from 12 to 107 per 
10,000,000 population in the ASALs and from 29 to 121 in the non-ASALs. These 
are presented in figure 
Figure 4.5: Average number of equipment per 10,000,000 population 
in 2013 and 2018
Source of data: Health sector annual performance review report 2020/21
In both ASALs and non-ASALs, the drug availability rate was higher in private 
compared to public facilities for both periods 2013 and 2018. The number of 
health workers per 10,000 population per county also increased from an average 
of 13 in 2013 to 17 per 10,000 in 2018 for both ASAL and non-ASAL counties.
The Wilcoxon signed rank test results that test for the significance of the changes to 
the quality of healthcare after devolution indicate that the increase in the number 
of equipment were all significant after devolution for both ASALs and non-ASALs 
except for MRI machines which increased marginally. Other aspects of healthcare 
quality like the availability of drugs based on the percentage of individuals seeking 
OP healthcare in public health facilities who got all the drugs they needed at the 
facility reveal a significant decrease in the numbers for the period after devolution. 
The percentage of drug availability in private health facilities did not experience 
a significant change in the period after devolution. The number of healthcare 
workers increased significantly per county for the period under consideration.
24
Results and discussions
The quality of healthcare in the counties has improved indicated by an increase in 
the number of trained and skilled healthcare professionals and the number of major 
equipment like MRI scans, x-ray machines, ultrasound machines, and CT scans. 
This could be because devolution transferred the planning and service delivery 
function to the counties and thus counties have been able to employ more trained 
healthcare professionals (Council of Governors, 2017)(Council of Governors, 
2017)(Council of Governors, 2017). In 2015, the government entered a Managed 
Equipment Service (MES) with counties to outsource specialized healthcare 
equipment including theatre facilities, kidney dialysis machines, x-rays, assorted 
cancer machines, and ICU machines on behalf of the counties. The equipment 
was distributed to two hospitals per county and additional health facilities across 
the country bringing the total list of beneficiaries to 119 facilities. This project 
was aimed at bringing healthcare closer to the people and reducing overreliance 
on national referral facilities (Ministry of Health, 2015). A report by IntraHealth 
International (2015) indicates that between 2013 and 2015, 7,484 health workers 
were recruited in the counties. On the other hand, many ASAL counties have not 
been able to achieve the WHO standards of 23 trained health personnel per 10,000 
indicative of a gap that still needs to be filled. These findings are supported by a 
systematic review study by Masaba et al., (2020) who also note that some counties 
are still understaffed in the WHO-recommended number of doctors and nurses 
per 10,000 population. The number of major healthcare equipment per county is 
also still low and more funds could be allocated for the purchase and maintenance 
of the existing ones. Drug availability shown by the percentage of patients who 
got all or some of the drugs prescribed to them decreased in the public healthcare 
facilities across both Asal and non-Asal counties, this could be attributed to drug 
stockouts Toroitich et al., (2022) due to the dues owed to KEMSA by counties, 
the main drugs supplier to public health facilities. The availability in the private 
health facilities was however found to be higher and this could be attributed to the 
fact that they were not dependent on supply from KEMS.
25
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
5. Conclusion 
This study aimed to find the effect of devolution on healthcare service delivery 
across counties. Different counties have shown mixed results with the rate 
of improvement being slower in ASALs especially the marginalized counties 
compared to non ASAL counties. Whereas improvements have been observed 
in various health facilities in terms of an increase in the number of equipment 
and infrastructure and an increase in the number of healthcare workers, many 
counties have not achieved the WHO recommended ratios indicating an existing 
gap that needs to be improved. 
5.1 Policy Recommendations 
i) Quality 
To increase drug availability in the public facilities, there is need to encourage the 
counties that have not passed the facility improvement fund bill to fast track its 
adoption. This allows all counties the autonomy to retain locally collected revenue, 
provide medical supplies, and improve the quality of healthcare. Additionally, 
counties can implement and strengthen quality programs involving regular 
assessments, monitoring and evaluation. 
ii) Accessibility
To address the issue of catastrophic health expenditure which increased in the 
period after devolution, there is need for ministry of health and county health 
department to educate the public on the importance of prioritizing preventive 
health services. This could contribute to early detection of diseases and bring 
down the financial burden of curing them. Additionally, counties can utilise ICT 
innovations as stipulated on eHealth Policy which will enhance access and depart 
from traditional healthcare delivery.
iii) Acceptability
To address the reduction in the number of people who expressed satisfaction with 
the healthcare received in public and private facilities, there is need for training 
health workers in the national and county health facilities on the importance of 
maintaining highest confidentiality levels when handling patients and keeping 
their data. In addition, the data collected should be used only for the intended 
purpose and as outlined in the law.
iv) Availability
To address the limited number of qualified healthcare professionals and bed 
facilities in some counties and ensure equal access to health across the country, 
there is a need for these counties to embrace the digitalization of the healthcare 
26
Conclusion
system which includes telemedicine and telehealth. Steps towards achieving this 
include fast-tracking the implementation of the digital health bill passed by the 
cabinet in August 2023 which obligates the national and county governments to 
establish and fully equip e-health centres to accelerate the uptake of e-health. This 
can be achieved through the partnership of county health departments and digital 
enablers like Safaricom. 
27
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
References
Abimbola, S., Baatiema, L., & Bigdeli, M. (2019). The impacts of decentralization 
on health system equity, efficiency and resilience: A realist synthesis of the 
evidence. In Health Policy and Planning (Vol. 34, Issue 8, pp. 605–617). 
Oxford University Press. https://doi.org/10.1093/heapol/czz055
Bossert, T. (1998). Decentralization of health systems: decision space, innovation 
and performance. Social Science Med, 47(10), 1513–1527.
Commission for Constitution Implementation. (n.d.). Integrating Constitutional 
Values and Principles and the Rights Based Approach in the Operational 
Frameworks of County Health Care Facilities 2015 Assessment Report for 
Kitui County Assessment Report for Kitui County 2015.
Costa-Font, J., & Perdikis, L. (2018). “Systems or Federacies”?
Council of Governors. (2017). ANNUAL STATUTORY REPORT.
Erk, J. (2014). Federalism and Decentralization in Sub-Saharan Africa: Five 
Patterns of Evolution. Regional and Federal Studies, 24(5), 535–552. 
https://doi.org/10.1080/13597566.2014.971769
Government of Kenya. (2020). Report by AD-HOC Committee to investigate the 
managed Equipment Services.
Government of Kenya. (2022). Thirteenth report on status of the equalization 
fund.
Government of Kenya (GoK). (2016). Devolution and Public Participation In 
Kenya Civic Education Handouts for Participants.
Hémet, C., Wren-Lewis, L., Mahoney, J., Jourdan, B., Banerjee, A., Behaghel, L., 
Cogneau, D., Crespin-Boucaud, J., Duflo, E., ere, N., Lambert, S., Macours, 
K., Michel, B., Suri, T., Vanden Eynde, O., & Zhuravskaya, E. (2023). 
Decentralization, Ethnic Fractionalization, and Public Services: Evidence 
from Kenyan Healthcare.
IntraHealth International. (2015). Kenya makes progress toward hrh commitments 
in a devolved health system. www.intrahealth.org
Kairu, A., Orangi, S., Mbuthia, B., Ondera, J., Ravishankar, N., & Barasa, E. (2021). 
Examining health facility financing in Kenya in the context of devolution. 
BMC Health Services Research, 21(1). https://doi.org/10.1186/s12913-
021-07123-7
Khaunya, M. F., Wawire, B. P., & Chepng’eno, V. (2015). International Journal 
of Economics, Finance and Management Devolved Governance in Kenya; 
Is it a False Start in Democratic Decentralization for Development? 4(1). 
http://www.ejournalofbusiness.org
Mackenzie, F., & Taylor, D. R. F. (1987). District Focus as a Strategy for Rural 
Development in Kenya: The Case of Murang’a District, Central Province. 
Canadian Journal of Development Studies / Revue Canadienne d’études 
28
References
Du Développement, 8(2), 299–316. https://doi.org/10.1080/02255189.19
87.9670192
Maina, T. (2015). CATASTROPHIC HEALTH EXPENDITURES AND 
IMPOVERISHMENT IN KENYA.
Makokha, R. O., & Amis, P. (2017). Does Decentralization Improve the Provision 
of Health Services? Evidence from Kisumu and Makueni Counties in 
Kenya. University of Birmingham.
Martín, S., Quintana, B., & Barrientos, D. (2016). Wilcoxon signed-rank-based 
technique for the pulse-shape analysis of HPGe detectors. In Nuclear 
Instruments and Methods in Physics Research, Section A: Accelerators, 
Spectrometers, Detectors and Associated Equipment (Vol. 823, pp. 32–
40). Elsevier B.V. https://doi.org/10.1016/j.nima.2016.03.094
Masaba, B. B., Moturi, J. K., Taiswa, J., & Mmusi-Phetoe, R. M. (2020). Devolution 
of healthcare system in Kenya: progress and challenges. In Public 
Health (Vol. 189, pp. 135–140). Elsevier B.V. https://doi.org/10.1016/j.
puhe.2020.10.001
Mbuthia, B., Vîlcu, I., Ravishankar, N., & Ondera, J. (2019). Purchasing at the 
county level in Kenya. www.thinkwell.global
McCollum, R., Limato, R., Otiso, L., Theobald, S., & Taegtmeyer, M. (2018). 
Health system governance following devolution: Comparing experiences of 
decentralisation in Kenya and Indonesia. BMJ Global Health, 3(5). https://
doi.org/10.1136/bmjgh-2018-000939
Ministry of Health. (2015). Demystifying the Managed Equipment Services (MES) 
project in Kenya.
Nyandiko, N. O. (2020). Devolution and disaster risk reduction in Kenya: 
Progress, challenges and opportunities. International Journal of Disaster 
Risk Reduction, 51. https://doi.org/10.1016/j.ijdrr.2020.101832
Pietrzyk, S., Moonzwe, L., Nzinga, J., Agweyu, A., D MacQuarrie, K. L., Prosnitz, 
D., & Schmale, A. (2018). Place of Delivery: Perceptions, Tensions, and 
Experiences Results from a Study in.
RESSYT. (2016). CASE STUDIES FOR HEALTH SYSTEMS AND POLICY 
ANALYSIS CASE STUDY OF GOVERNANCE CHANGE Health system 
decentralisation in Kenya. http://hpsa-africa.org/
Rey, D., & Neuhäuser, M. (n.d.). Wilcoxon Signed Rank Test.
Rosner, B., Glynn, R. J., & Lee, M. L. T. (2006). The Wilcoxon signed rank test for 
paired comparisons of clustered data. Biometrics, 62(1), 185–192. https://
doi.org/10.1111/j.1541-0420.2005.00389.x
Scheff, S. W. (2016). Nonparametric Statistics. In Fundamental Statistical 
Principles for the Neurobiologist (pp. 157–182). Elsevier. https://doi.
org/10.1016/B978-0-12-804753-8.00008-7
29
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Toroitich, A. M., Dunford, L., Armitage, R., & Tanna, S. (2022). Patients Access to 
Medicines – A Critical Review of the Healthcare System in Kenya. In Risk 
Management and Healthcare Policy (Vol. 15, pp. 361–374). Dove Medical 
Press Ltd. https://doi.org/10.2147/RMHP.S348816
Wanyande, P., & Mboya, T. (n.d.). DEVOLUTION: THE KENYA CASE.
Wanzala, M. N., & Oloo, M. O. (2019). Role of Devolution in Health Service 
Delivery, Health Workforce and Medical Commodities Acquisition: A 
Review. Saudi Journal of Medicine Abbreviated Key Title: Saudi J Med. 
https://doi.org/10.36348/sjm.2019.v04i07.004
World Bank Group. (2022, October 6). Universal Health Coverage. Https://Www.
Worldbank.Org/En/Topic/Universalhealthcoverage.
30
Appendix
Accessibility
Distance covered to reach an outpatient health facility before and after devolution
Figure 4.1 shows the average distance in kilometres covered by an individual to reach 
an outpatient health facility where they seek care from. On average, individuals 
across the counties covered longer distances in 2018 as compared to 2013 in the 
exception of Laikipia, Baringo, Siaya, Bomet, Uasin Gishu, and Murang’a counties 
where the average distance covered after devolution was shorter. However, the 
recommended distance of 5 kilometres has only been achieved in Uasin Gishu, 
Migori, Homabay, Busia, and Murang’a in 2018.
Physical access to inpatient facilities based on facility type
Average distance to an IP private health facility before and after devolution per 
county
The distance covered to reach an inpatient private health facility increased in 
most of the arid counties except for Isiolo. In the semi-arid counties however, the 
distance has reduced.
31
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
Average distance to an IP county health facility in 2018 and government facility 
in 2013
On average, the distance to an inpatient county health facility has reduced across 
the non-ASAL counties, this is not the case in some of the arid and semi-arid 
counties including Samburu, Marsabit, Kajiado, Makueni, Kitui, Laikipia, Baringo 
and Kericho where the average distance to county facilities increased after 
devolution.
Out of pocket expenditure in a government health centre before and after 
devolution
32
On average, the out-of-pocket expenditure by individuals to access health 
increased in most of the counties except Uasin Gishu, Tharaka-Nithi, Lamu and 
Kisumu.
Out of Pocket Expenditure for Accessing healthcare from county referral hospital 
in 2018 and government hospital in 2013
The out-of-pocket expenditure for the county referral hospitals display mixed 
results, whereas the cost increased in some counties such as Isiolo, Kajiado, 
Nakuru, Trans Nzoia, and Nyamira, it decreased in the remaining counties. 
Time taken to reach an outpatient health facility of choice per county in minutes
The trend across the counties shows that there has been an increase in the number 
of minutes taken to reach an outpatient health facility in all counties except for 
Bungoma. Comparatively, the average time taken to reach health facilities in 
arid counties is not very different from the average time taken to reach facilities 
in semi-arid and non-ASAL counties. In all the counties however, the WHO 
recommended time of 30 minutes has not been exceeded in any of the counties. 
This is an indication of the increased distribution of the outpatient health facilities 
33
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
across the counties in a way that they can be reached by patients within 30 minutes 
regardless of the means of transport used.
Availability
Number of beds per 10,000 population before and after devolution
On average, there has been an increase in the number of beds in the health facilities 
in most of the counties except for Tana River, Garissa, Kwale, Kilifi, Taita Taveta 
and Lamu counties. 
Number of operational health facilities per county before and after devolution per 
10,000 people
There has been an increase in the number of operational health facilities per 
county in 2018 as compared to 2013. However, the number of facilities in arid 
34
counties is lower compared to semi-arid and non-ASAL counties. 
Acceptability
Appendix Figure 9: Privacy in outpatient facilities before and after devolution
Figure 4.4 shows the average number of people who were satisfied with the level 
of privacy received in the healthcare facility visited. A decreasing trend is observed 
across both ASAL and non-ASAL counties for the period after devolution in most 
of the counties except Samburu, Marsabit, Turkana, Kilifi, Nyeri, Busia, and Uasin 
Gishu.
Appendix Figure 10: Number of health workers per 10,000 population before and 
after devolution.
Figure 4.8 shows the number of health workers per 10,000 for every county for 
the period 2013 and 2018 respectively. Although there has been an increase in the 
number of health workers in all the counties, only seven counties have achieved 
35
An assessment of the effect of devolution on the healthcare service delivery in ASAL counties of Kenya
the WHO recommendation of 23 healthcare workers per 10,000 people. 
Conditional grants from the national government include Level 5 hospital funds, 
hospital user fees forgone by county governments, universal health care, free 
maternal health care, and leasing of medical equipment. Conditional grants have 
ranged between KES 2 Billion to 23 Billion between the Financial Years 2013/14 
and 2021/22, 2021/22 being KES 8 Billion.
36