PTA report 2016March30 FINAL

FINAL REPORT
March 2016
POST-TRAINING ASSESSMENT OF
HEALTH WORKERS TRAINED UNDER
THE FUNZOKENYA PROJECT

Post-Training Assessment of Health Workers
Trained under the FUNZOKenya Project: Final Report ii
TABLE OF CONTENTS
LIST OF ACRONYMS .........................................................................................................................................iii
EXECUTIVE SUMMARY .....................................................................................................................................iv
INTRODUCTION .................................................................................................................................................1
BACKGROUND...............................................................................................................................................................1
THE FUNZOKENYA PROJECT ...........................................................................................................................................1
METHODS............................................................................................................................................................4
ASSESSMENT SITES AND PARTICIPANTS ..............................................................................................................................4
DATA COLLECTION METHODS ..........................................................................................................................................4
DATA COLLECTION PROCEDURES.......................................................................................................................................7
DATA MANAGEMENT AND ANALYSIS .................................................................................................................................8
FINDINGS ............................................................................................................................................................9
TRAINING APPROACH .....................................................................................................................................................9
REPORTED EFFECTS OF TRAINING ON SERVICE DELIVERY......................................................................................................20
DISCUSSION......................................................................................................................................................33
RECOMMENDATIONS .....................................................................................................................................36
CONCLUSIONS..................................................................................................................................................37
REFERENCES......................................................................................................................................................38
APPENDIX 1. SUPERVISOR INTERVIEWS......................................................................................................41
APPENDIX 2. KEY INFORMANT INTERVIEWS..............................................................................................42
APPENDIX 3. FOCUS GROUP DISCUSSIONS ................................................................................................43
APPENDIX 4. HEALTH WORKER INTERVIEWS.............................................................................................44
APPENDIX 5. OBSERVATIONS OF CLIENT INTERACTIONS........................................................................45

Trained under the FUNZOKenya Project: Final Report iii
LIST OF ACRONYMS
APHIAPlus AIDS, Population and Health Integrated Assistance Plus
ART Antiretroviral therapy
CASCO County AIDS and STI coordinator
CCC Comprehensive care clinic
CHD County health director
CHMT County health management team
CME Continuing medical education
DRH Division of Reproductive Health
EmONC Emergency obstetric and newborn care
ENC Essential newborn care
FBO Faith-based organization
FGD Focus group discussion
FP Family planning
HRD Human Resource Department
HRH Human resources for health
HTC HIV testing and counseling
IMCI Integrated Management of Childhood Illness
IYCF Infant and young child feeding
KEMRI Kenya Medical Research Institute
KMTC Kenya Medical Training College
LARC Long-acting reversible contraception
LSTM Liverpool School of Tropical Medicine
MNCH Maternal, neonatal, and child health
MOH Ministry of Health
MTC Medical training college
NASCOP National AIDS and STI Control Program
PEPFAR President’s Emergency Plan for AIDS Relief
PMTCT Prevention of mother-to-child transmission of HIV
PPH Postpartum hemorrhage
PTA Post-training assessment
RMHSU Reproductive and Maternal Health Services Unit
STI Sexually transmitted infection
TOT Training-of-trainers
USAID United States Agency for International Development
USG United States government

Trained under the FUNZOKenya Project: Final Report iv
EXECUTIVE SUMMARY
Background
In Kenya, the health system has long been characterized by a serious shortage of human
resources for health (HRH). Providing high-quality education and training for new and existing
health workers is a recognized strategy to respond to the HRH crisis, provide health workers
with the knowledge and competencies that they need to provide responsive, effective, and
comprehensive health services and, ultimately, improve service delivery and quality of care in
key areas such as HIV prevention, care, and treatment. Evidence suggests that improving the
quality of care can improve health outcomes.
The FUNZOKenya project, led by IntraHealth International and funded by the President’s
Emergency Plan for AIDS Relief through the United States Agency for International
Development, aims to transform health workforce training systems by providing support to the
Ministry of Health (MOH) and Government of Kenya. FUNZOKenya addresses four areas of
health workforce strengthening: (1) increase the number of new health workers trained; (2)
support health workers’ training needs through in-service training; (3) strengthen capacity of
medical training institutions; and (4) strengthen mechanisms for relicensure and accreditation.
This report focuses on the in-service training component.
In 2013, FUNZOKenya introduced an in-service training strategy to establish regional in-service
training hubs and enhance collaboration between academia and health practitioners. A
competitive process selected eight regional training hubs led by 14 medical training colleges or
universities and including practicum sites. The institutions included Kenya Medical Training
College (Garissa, Nairobi, and Nakuru campuses); Kenya Methodist University; Kenyatta
University; Kijabe School of Nursing; Maseno University School of Medicine; Masinde Muliro
University of Science and Technology; Maua Methodist Hospital School of Nursing; Mission for
Essential Drugs and Supplies; Moi University College of Health Sciences; Outspan Medical
College; Pwani University; and Tenwek School of Nursing.
FUNZOKenya supported MOH technical departments to train and certify faculty at the training
institutions in three priority areas: (1) HIV/AIDS, including prevention of mother-to-child
transmission (PMTCT), antiretroviral therapy (ART), and HIV testing and counseling (HTC); (2)
maternal, neonatal, and child health (MNCH), including emergency obstetric and neonatal care
(EmONC), Integrated Management of Childhood Illness (IMCI), and infant and young child
feeding (IYCF); and (3) family planning (FP), including long-acting reversible contraceptives
(LARCs) such as Implanon NXT.
The training institutions collaborated with county health management team (CHMT) members
and the USAID-supported APHIAPlus projects that support specific health facilities on service
delivery. CHMT members were responsible for identifying and preparing training practicum sites
and materials and identifying health workers for training, in consultation with APHIAPlus and
supervisors.

Trained under the FUNZOKenya Project: Final Report v
FUNZOKenya Training Strategy Highlights
 Identified eight regional training hubs
 Through a competitive process, selected 14 training institutions capable of delivering in-service training to
standards acceptable to Ministry of Health technical departments
 Built training institution capacity for in-service training in key HIV, MNCH, and FP technical areas
 Enhanced collaboration between practitioners, academia-based trainers, CHMTs, and APHIAPlus.
Trainers included certified faculty from the regional training institutions, county-level MOH
master trainers, and APHIAPlus technical experts, who together trained selected health workers
in accordance with MOH-approved curricula, policies, and guidelines. Trainees received course
materials, meals and refreshments, and reimbursement for public transportation. Participants
located far from the training institutions also received accommodation. No additional
allowances or per diems were provided.
Between 2011 and 2015, FUNZOKenya supported the in-service training of almost 12,000 health
workers using the regional training approach. Between October 2013 and September 2014 (the
focus of this assessment), FUNZOKenya trained 2,898 health workers in HIV/AIDS, MNCH, and
FP services. In 2015, FUNZOKenya carried out a post-training assessment (PTA) to:
 Elucidate the advantages and disadvantages of its training approach
 Describe perceived effects of the in-service trainings on aspects of service delivery.
Methods
FUNZOKenya carried out the PTA in August 2015 in ten purposively selected counties out of the
47 counties that accessed the FUNZOKenya-supported regional in-service training approach.
The ten counties (Baringo, Homabay, Kilifi, Kitui, Mombasa, Nakura, Narok, Tharaka-Nithi, Trans-
Nzoia, and Vihiga) were chosen on the basis of high HIV burden, number of health workers
trained in five focus areas (ART, HTC, PMTCT, EmONC, and FP/LARC), and other donor-relevant
criteria, with a focus on health workers trained from October 2013 through September 2014. The
PTA elicited multiple perspectives to develop a rich understanding of the training approach,
health workers’ post-training knowledge and performance, and the health system context
shaping quality of care and delivery of services after training.
The PTA used a mix of qualitative and quantitative data collection methods, including several
methods involving health workers trained with FUNZOKenya support. The five methods included
semi-structured and in-depth interviews with 41 facility supervisors and 49 national and county-
level key informants; ten focus group discussions with trained health workers; structured
interviews with 222 purposively selected trained health workers; structured observations of 86
trained health workers and 138 client interactions; and structured exit interviews with 138 clients
post-observation.

Trained under the FUNZOKenya Project: Final Report vi
Qualitative data were coded using an iterative approach and analyzed in NVivo. Semi-structured
and quantitative data were entered and analyzed in SPPS, Stata, and Excel. Ethical approval was
obtained from the Kenya Medical Research Institute Ethics Review Committee and IntraHealth,
and all participants provided informed consent before responding to questions or starting
structured observations.
Limitations of the PTA include a small number of observations of health worker-client
interactions for some services (e.g., EmONC and FP/LARC), lack of a comparison group or
comparison data, and difficulties in differentiating the influence of FUNZOKenya-supported
trainings on observed results from the potential effects of other training initiatives or health
systems support activities.
Results
Training Approach
County leaders and trainees shared positive comments about the FUNZOKenya training
approach, describing the training institutions as “dynamic” and “credible.” Key informants from
partner organizations had more mixed opinions. Nonetheless, training institution and partner
organization representatives gave the FUNZOKenya approach high marks for building capacity
at the training institutions, describing capacity building as the “number one benefit.”
Assessment participants identified other advantages of the training model as well as some
disadvantages, summarized below.
Key Advantages and Disadvantages of FUNZOKenya-Supported Training Approach
Advantages
Planning “Consultative” and “collaborative” process to plan trainings and select trainees
Materials Training manuals from approved curricula in keeping with established guidelines and standards
Trainers Qualified and “seasoned” trainers at training institutions who are “used to teaching”
Selection Counties adept at “balancing” selection criteria and addressing gaps without favoritism
Sustainability Training institutions not constrained by donor funding cycles and willing to innovate
Disadvantages
Sessions Duration not always sufficient to cover complex material fully (e.g., ART)
Venue Training institution locations not always convenient for attendees traveling from home each day
Trainers University lecturers lack “hands-on” experience that is relevant to health workers (e.g., EmONC)
Practice Inadequate clinical practice component
Motivation Institution-based environment and lower allowances “demoralizing” (vs. “extra comfort” of hotels)
Recognition Certificates not delivered in timely manner
Service Delivery
Most trainees, supervisors, and key informants described improvements in availability and
quality of services following the FUNZOKenya-supported trainings. Nearly all interviewed health
workers (96%) reported that, post-training, they were providing the services for which they had
been trained. For example, most (77%) health workers trained in EmONC reported attending a
delivery in the previous week of the assessment period. Health workers trained in EmONC (53%)
and FP/LARC (45%) were especially likely to report providing additional or improved services
post-training. Service delivery improvements cited by health workers in focus group discussions

Trained under the FUNZOKenya Project: Final Report vii
included better integration of services and reduced reliance on referrals. Key informants and
supervisors shared trainees’ perceptions concerning the positive effects of FUNZOKenya-
supported training, describing improved health worker attitudes and motivation, increased
availability and uptake of health services, and even improved clinical outcomes. Key informants
reported that the trainings had resulted in more up-to-date and “standardized” care, better
adherence to clinical guidelines, and a higher quality of care that eliminates “guesswork.”
Supervisors noted that enhanced quality of care and counseling for HTC, PMTCT, and ART, in
particular, had translated into better uptake and adherence.
Client Experiences
Many health workers perceived that the trainings had improved experiences for their clients.
Over half (56%) of surveyed health workers reported that the trainings had led to reduced wait
times, and 62% reported improved protection of confidentiality. Approximately 5% of surveyed
trainees noted that the trainings had improved their counseling of clients. Health workers also
brought up these positive benefits for clients in the focus group discussions. Supervisors
commented on waiting times but qualified their remarks by pointing out that staff shortages
limited the extent to which waiting times could be reduced. Some supervisors suggested that
waiting times had actually increased because their facilities were offering higher quality of care
and, as a result, attracting more clients. Clients rated their clinical experience with health workers
trained with FUNZOKenya support as very positive, with over 80% reporting that their
experience was either “excellent” or “good.” Most clients (79%) also rated wait times as
“adequate” rather than “too long.”
Health Worker Knowledge and Practices
Observations of health workers’ practices demonstrated that most health workers who had
received FUNZOKenya-supported training were following HTC and PMTCT testing and
counseling protocols. Only two HTC areas needed improvement: establishing what prompted
the visit for HIV testing (completed in only 29% of observed interactions), and reading the test
results within 15 minutes (64%). Health workers who tested expectant mothers during antenatal
care/PMTCT sessions provided information on the importance and benefits of HIV testing and
obtained informed consent (79% of observed interactions), and, in post-test counseling,
assessed readiness to receive results (74%), discussed prevention of future infections (78%), and
explored and encouraged partner testing (95%).
Observations of trained health workers’ practices showed variable completion of the steps in the
management of returning ART clients. In the majority of encounters, health workers assessed the
client’s social support (67%), chief complaints and medical history (58%), and weight (79%).
However, health workers conducted TB screening in less than half (44%) of observed encounters
and used laboratory assessments to monitor ART or assess ART-related toxicities in less than
one-third. On the other hand, in approximately two-thirds of client interactions, trained health
workers monitored ART efficacy (60%), managed ART side effects (67%), and developed or
monitored adherence treatment plans (67%).

Trained under the FUNZOKenya Project: Final Report viii
The PTA was not able to observe enough specific types of obstetrical and newborn client
interactions to assess health workers’ EmONC practices. However, EmONC-trained health
workers were interviewed about knowledge of signs and actions for postpartum hemorrhage
(PPH) and obstructed labor. The majority correctly identified three signs of PPH, although only
27% identified all PPH signs. Similarly, only 20% of health workers correctly identified all
appropriate PPH management actions, although the majority correctly identified five actions.
Health workers’ knowledge of the signs of obstructed labor was more mixed. The majority
correctly identified four signs of obstructed labor, although only 7% identified all signs. Only 9%
of health workers correctly identified all actions for obstructed labor.
Health System Constraints
Despite positive perceptions of the trainings and their effects, respondents at all levels
acknowledged that systemic health systems challenges often affected health workers’ ability to
apply skills acquired during training. The most common health system constraint, cited by over a
third (37%) of health workers surveyed, was inadequate staffing, and 16% cited heavy workloads.
Health workers trained in HTC (and assumed to be providing HTC services) were more likely
(79%) to report staff shortages as a major service delivery challenge, while health workers
trained in FP/LARC were least likely (18%). The second most common constraint preventing
trainees from providing services according to training competencies was stockouts (29%),
followed by lack of equipment (20%) and inadequate infrastructure (15%). HTC-trained health
workers were most likely to cite stockouts as an obstacle to providing services (70%), followed
by health workers trained (and presumably providing services) in FP/LARC (36%).
Discussion
The post-training assessment produced robust data to elucidate the strengths and weaknesses
of the in-service training approach introduced by FUNZOKenya that has trained almost 12,000
health workers over a four-year period. The assessment’s findings suggest that the regional
training approach holds considerable promise. One of the most salient advantages identified
was the capacity built at the eight regional training institutions and the training model’s implied
potential for longer-term sustainability. Some health workers participating in FUNZOKenya-
supported trainings were dissatisfied with the training logistics, citing inconvenient locations of
the training venues or “demoralizing” allowances and venue quality. Altering trainees’
expectations regarding “rewards” for participating in in-service training may take time. Despite
these disadvantages, many health workers and supervisors requested expanded training.
Respondents’ feedback also pointed to the importance of providing post-training follow-up,
appropriate supervision, further on-the-job training, and mentorship to ensure that health
workers master and maintain new competencies and are able to put them into practice for
improved quality of services.
The PTA’s design did not allow for meaningful quantitative assessment of the complex
relationship between training and service delivery outcomes. However, the rich qualitative data
generated by the assessment indicate that training participants, their supervisors, and officials at
the county level and beyond all perceived numerous beneficial effects of the FUNZOKenya-

Trained under the FUNZOKenya Project: Final Report ix
supported trainings on a variety of intermediate effects, both subtle and straightforward,
including at the client, health worker, and facility levels. Moreover, observations and survey
items identified strong knowledge and many strong practices among health workers who had
received FUNZOKenya-supported training in HTC, PMTCT, and ART. The overwhelming majority
(often 100%) were observed to be following infection control practices and HIV pretesting and
post-test counseling protocols. Observations of trained health workers’ management of
returning ART clients found that, in most encounters (58%-79%), health workers assessed the
client’s condition, monitored ART efficacy, managed ART side effects, and developed or
monitored adherence treatment plans. Interviews with EmONC-trained health workers identified
generally high knowledge levels for postpartum hemorrhage (comparable to or higher than
levels measured in a 2010 national health facility survey in Kenya). The PTA sample’s knowledge
of the signs of obstructed labor was more mixed, suggesting that there may be room for further
improvement in some areas of EmONC training.
A number of recommendations can be gleaned from the PTA findings:
1. Expand the overall number of training slots available.
2. Schedule trainings more frequently.
3. Provide training on neglected topics such as chronic disease management and supply
chain management.
4. Provide routine refresher training and on-the-job training to accommodate evolving
clinical guidelines.
5. Develop and adhere to a consistent system to assess health workers’ training needs and
target training slots to the participants and facilities most in need of training support.
Full implementation of iHRIS Train and its integration with iHRIS Manage at the county
and national levels could facilitate appropriate trainee selection, redeployment, and
retention.
6. Ensure that supervisors receive up-to-date training in the areas in which they are
responsible for supervision.
7. Work toward greater coordination of training approaches across counties by the MOH
and training partners, particularly with regard to allowances, venues, and other logistic
aspects of training.
8. Include a mix of trainers representing both academic and clinical settings.
9. Build the “hands-on” training capacity of trainers from the regional training
institutions.
10. Include more practical training, preferably in clinical settings, recognizing that
“practicing on each other isn’t good enough.”
11. Ensure that completion-of-training certificates are issued without delay, preferably
online, following training.

Trained under the FUNZOKenya Project: Final Report x
12. Strengthen follow-up and provide ongoing mentorship for trainees to ensure solid
skills and performance before skills are passed on to co-workers.
13. Enhance county health departments’ ability to address health system constraints (such
as stockouts, lack of equipment, and staff shortages) and their effects on post-training
application of skills learned.
14. At the front end, consider measurement strategies to assess linkages between in-
service training and client, health systems, and population-level outputs and outcomes.
The findings outlined in this report can inform key stakeholders about gaps and needed
improvements in training approaches and the health system supports necessary to ensure
health workers can use learned competencies to strengthen service delivery. The lack of a
comparison group precludes any conclusive determination regarding the relative influence of
FUNZOKenya-supported training on the quality and quantity of services delivered by trainees or
its eventual impact on health outcomes. Nonetheless, the varied evidence collected, including a
rich assortment of qualitative data and triangulation of different perspectives, points to
potentially meaningful effects on health worker confidence, skills, and adherence to clinical
protocols (particularly for HIV), with perceived spillover effects that include higher-quality
interactions between trained health workers and clients and greater availability and uptake of
key services.

Trained under the FUNZOKenya Project: Final Report 1
INTRODUCTION
Background
Numerous countries face significant health workforce challenges, including a shortage of trained
and qualified health workers to provide needed services (Chen et al. 2004). In Kenya, the health
system has long been characterized by a serious shortage of human resources for health (HRH)
(Ministry of Health 2014). Considering Kenya’s vital nursing cadre alone, there is a countrywide
shortage of nurses in the public sector, exacerbated by maldistribution of nurses across rural
and urban counties (Wakaba et al. 2014). Health outcomes tend to be poorer where health
worker density is lower (Anand and Bärnighausen 2004; Wakaba et al. 2014).
Availability of skilled and confident providers is critical for quality service provision (Bailey et al.
2013). Providing high-quality education and training for new and existing health workers is a
recognized strategy to respond to the HRH crisis and improve service delivery and quality of
care in key areas such as HIV prevention, care, and treatment (O’Malley, Perdue, and Petracca
2013). In-service training can reinforce and update health workers’ knowledge and can improve
the quality of services offered to clients if effective educational techniques are employed
(Bluestone et al. 2013), such as adequate numbers of properly skilled instructors and locally
adapted training materials (Kawaguchi and Mori 2010). The effectiveness of in-service training
and other interventions to improve health workers’ performance and service quality for clients is
also dependent on other health systems factors, such as ongoing supervision and management,
available supplies and materials, and adequate staff with the requisite skills and sufficient
motivation (Blacklock et al. 2016).
Kenya’s National Human Resources for Health Strategic Plan 2009–2012 identified weaknesses in
both preservice education and in-service training (Ministry of Medical Services 2009), and a
performance needs assessment found that in-service training did not always correlate with
improvements in service delivery (Ministry of Health 2011). High-quality training that focuses on
improving service delivery is essential to furnish health workers with the knowledge and
competencies that they need to provide responsive, effective, and comprehensive care in Kenya.
The FUNZOKenya Project
The FUNZOKenya project, led by IntraHealth International and funded by the President’s
Emergency Plan for AIDS Relief (PEPFAR) through the United States Agency for International
Development (USAID), aims to transform health workforce training systems by providing
support to the Ministry of Health (MOH), Government of Kenya, county-level stakeholders,
private sector providers (including faith-based organizations [FBOs]), and other US government
(USG) implementing partners and stakeholders. FUNZOKenya addresses the four areas of health
workforce strengthening listed below. This report focuses on the in-service training components
(#2 and #3).

1. Boost the number of new health workers trained by increasing the admissions capacity
of preservice education institutions and expanding access to loans and scholarships for
students in the health professions.
2. Support current health workers’ training needs through in-service training.
3. Strengthen the training capacity of medical training institutions through curriculum
review and development, faculty development, and upgrading of training management
systems and infrastructure.
4. Link in-service training to health worker relicensure and increase the capacity of
regulatory bodies and councils to accredit and monitor training.
FUNZOKenya-Supported In-Service Training Activities
In 2013, FUNZOKenya introduced an in-service training strategy with the aim of establishing
regional in-service training hubs and enhancing collaboration between academia and health
practitioners. The eight regional training hubs are led by a medical training college or university
and include corresponding practicum sites for hands-on training. A competitive process
identified 14 institutions that had the capacity to deliver training to standards acceptable to
MOH technical departments such as the National AIDS and STI Control Programme (NASCOP)
and the Reproductive and Maternal Health Services Unit (RMHSU). The institutions included
Kenya Medical Training College (Garissa, Nairobi, and Nakuru campuses); Kenya Methodist
University; Kenyatta University; Kijabe School of Nursing; Maseno University School of Medicine;
Masinde Muliro University of Science and Technology; Maua Methodist Hospital School of
Nursing; Mission for Essential Drugs and Supplies; Moi University College of Health Sciences;
Outspan Medical College; Pwani University; and Tenwek School of Nursing.
FUNZOKenya supported MOH technical departments to train and certify faculty at the training
institutions in three priority areas:
 HIV/AIDS, including prevention of mother-to-child transmission (PMTCT), antiretroviral
therapy (ART), and HIV testing and counseling (HTC)
 Maternal, neonatal, and child health (MNCH), including emergency obstetric and
neonatal care (EmONC), Integrated Management of Childhood Illness (IMCI), and infant
and young child feeding (IYCF)
 Family planning (FP), including long-acting reversible contraceptives (LARC) such as
Implanon NXT.
The training institutions collaborated with county health management team (CHMT) members
and the USAID-supported APHIAPlus projects that support specific health facilities on service
delivery. CHMT members identified and prepared practicum sites (such as hospitals,
comprehensive care clinics [CCCs], and maternity wards) and materials such as test kits and
Implanon NXT. The CHMTs also identified health workers for training, in consultation with
APHIAPlus and facility in-charges.

Trainers included certified faculty from the regional training institutions, county-level MOH
master trainers, and APHIAPlus technical experts (where applicable), who together provided
training in accordance with MOH-approved curricula, policies, and guidelines. Trainees received
course materials, meals and refreshments, and reimbursement for public transportation.
Participants located far from the training institutions also received accommodation. No
additional allowances or per diems were provided.
Since July 2012, FUNZOKenya has supported the in-service training of almost 12,000 health
workers. Between October 2013 and September 2014 (the period targeted by this post-training
assessment), FUNZOKenya supported training of almost 3,000 health workers in HIV testing,
treatment, and care, MNCH, and FP services (Table 1).
Table 1: Number of Health Workers Trained with FUNZOKenya Support, by Training Topic
Training topic
Number of health workers trained with FUNZOKenya
support
July ‘12-
Sept ‘12
Oct ‘12-
Sept ‘13
Oct ‘13-
Sept ‘14*
Oct ‘14-
Sept ‘15
TOTAL
Adult ART 185 185 776 0 1146
Cervical cancer screening 163 274 25 0 462
Commodity management 0 209 94 31 334
Clinical care of sexual
violence
0 55 0 0 55
EmONC 0 222 526 32 780
Essential newborn care (ENC) 27 0 0 0 27
Focused antenatal care 30 30
FP/LARC/contraceptive
updates
0 89 86 1805 1980
Harmonized HIV curriculum 0 107 0 0 107
HTC (New algorithm) 0 0 577 122 699
IMCI 53 71 59 0 183
IYCF 0 230 206 0 436
New ART guidelines 0 0 0 1218 1218
Pediatric HIV 0 1016 320 33 1369
Phlebotomy 0 0 15 0 15
PMTCT 267 1304 214 620 2405
TB management (HIV/TB
integration; pediatric TB;
multidrug-resistant TB)
87 84 0 119 290
GRAND TOTAL 812 3,846 2,898 3,980 11,536
*Target period for this post-training assessment

METHODS
In 2015, FUNZOKenya carried out a post-training assessment (PTA) to:
 Elucidate the advantages and disadvantages of its training approach
 Describe reported effects of the in-service trainings on aspects of service delivery.
The perspectives of policy-makers, county managers, trainers, partners, clinical supervisors,
health workers, and clients can all be valuable in pinpointing needed training and quality of care
improvements (Naidu 2009). Gathering information from these groups of stakeholders through
a mixture of quantitative and qualitative methods, the PTA sought to elicit multiple perspectives
to develop a rich understanding of the strengths and weaknesses of the training, health workers’
post-training performance, and the context shaping performance and quality of care.
Assessment Sites and Participants
The FUNZOKenya project works in 47 counties. The project purposively selected ten counties
(Baringo, Homabay, Kilifi, Kitui, Mombasa, Nakuru, Narok, Tharaka-Nithi, Trans-Nzoia, and
Vihiga) for inclusion in the PTA. The following criteria were used to select the ten counties:
1. Priority county as determined by PEPFAR or USAID
- High HIV burden (PEPFAR-focus counties)
- Scale-up of EmONC or FP services (USAID)
2. Type of trainings received (HTC, PMTCT, adult ART, EmONC, and/or FP/LARC)
3. Number of health workers trained
4. Geographic representativeness.
To select health workers, all providers who had participated in FUNZOKenya-supported trainings
in the five topic areas in the ten counties were listed by facility. Of these, the assessment
included health workers trained during the designated time period (October 2013 to September
2014) who were continuing to provide services in one of the five priority areas in the selected
facilities (n=222).
Data Collection Methods
The PTA applied a combination of quantitative and qualitative techniques and primary and
secondary data to investigate the assessment’s two primary questions. The mix of five methods
(listed below and summarized in Table 2) facilitated the triangulation of data to capture different
perspectives of the same phenomena, reveal more than one facet of a topic or problem, and
explore possible linkages between provider knowledge and performance and quality of care.

1. Semi-structured and in-depth interviews with national- and county-level stakeholders
and supervisors of trained health workers
2. Focus group discussions with trained health workers
3. Structured interviews with health workers
4. Structured observations of service delivery and provider-client interactions
5. Exit interviews with clients.
Table 2: Summary of PTA Data Collection Methods
Methods Participants Types of information collected Sample size
Semi-
structured and
in-depth
interviews
 Supervisors
 Key
informants
at national,
regional, and
county levels
 Opinions about the FUNZOKenya training
approach, including probes about the
advantages and challenges
 Selection of trainees and tracking of in-service
training
 Post-training support and supervision for health
workers
 Perceptions of effects of trainings on health
worker performance and service delivery
 41 supervisors
 49 key informants
-National experts
-Regulatory bodies
-Training institutions
-County stakeholders
Focus group
discussions
 Health
workers
1
 Opinions about FUNZOKenya training approach
 Systems in place for sharing information after
training
 Perceptions of effects of trainings on service
delivery
 Post-training support and supervision
 Challenges in providing services post-training
 10 focus groups
Structured
interviews with
health workers
 Health
workers
1
 Types of FUNZOKenya training received
 Services offered before and after FUNZOKenya
training
 Perceptions of improvements in service delivery
 Clinical knowledge related to the specific
training received
 222 health workers
Structured
observations
 Health
workers
serving
clients in five
key areas
1
 Health worker-client interactions, including
counseling and clinical aspects in each service
area (see Table 3)
 86 health workers
 138 interactions
Client exit
interviews
 Clients
exiting
structured
observations
 Satisfaction with received services
 Interactions with health workers during
consultation
 138 clients
1
Trained in HTC, PMTCT, adult ART, EmONC, or FP/LARC in the ten counties.
In-Depth and Semi-Structured Interviews
Semi-structured interviews were conducted with 41 supervisors from health facilities where
health workers had received FUNZOKenya-supported in-service training (see Appendix 1).
During the interviews, supervisors shared perceptions about the effects of the training on the
quality and availability of services in their facilities. Supervisors were also asked to share their

experiences related to selecting health workers for training and providing post-training support
and supervision.
In addition, 49 in-depth key informant interviews were conducted with national-level policy-
makers, regulatory bodies, training institutions, and, at the county level, CHMT members (see
Appendix 2). Key informants addressed the strengths and weaknesses of the FUNZOKenya-
supported in-service training and the perceived effects of the trainings on health worker
performance and service delivery.
Focus Group Discussions
The PTA conducted focus group discussions (FGDs) with health workers who had participated in
FUNZOKenya-supported in-service training in at least one of the five service delivery areas (see
Appendix 3). Each FGD included seven to ten participants and lasted approximately 45 minutes.
Health workers discussed their opinions about the FUNZOKenya training approach, their
perceptions of the effects of the training on their service delivery, and their post-training
experiences related to supervision, information sharing, and challenges in delivering services.
The discussions were held in health facility halls that were booked before the meetings. FGDs
were conducted with health workers in Baringo county (among those trained in EmONC and
PMTCT); Homabay, Tharaka-Nithi, and Vihiga counties (trained in adult ART); Kilifi county
(trained in EmONC and HTC); Mombasa county (trained in HTC); Nakuru county (trained in
PMTCT); and Narok county (trained in FP/LARC). No FGDs were conducted in Kitui or Trans-
Nzoia counties.
Structured Interviews with Health Workers
Structured interviews were conducted with a total of 222 health workers trained in the five focus
areas with FUNZOKenya support between October 2013 and September 2014 (see Appendix 4).
Health workers responded to questions about the types of training received with FUNZOKenya
support, attendance at trainings supported by other partners, types of services offered before
and after FUNZOKenya trainings, perceptions about improvements in service delivery related to
the trainings, and challenges in delivering services. The structured interviews also assessed
theoretical and clinical knowledge in HTC, PMTCT, adult ART, EmONC, and/or FP/LARC among
the sampled health workers who had received training in each of these areas.
Structured Observations
For each of the five service delivery areas, trained observers carried out structured observations
to assess the extent to which health workers (1) applied the knowledge and skills gained during
in-service training and (2) adhered to standards of care outlined in national MOH guidelines for
that particular type of service. Before initiating observations, data collectors introduced
themselves and sought the client’s permission to observe the service delivery interaction. The
observations were recorded on checklists tailored to the service being assessed. As a way of
providing direct feedback, observers subsequently reviewed the checklist with each health
worker. The checklists covered health worker behaviors and health worker-client interactions,
including counseling and health worker performance of specific procedures and examinations in
accordance with national guidelines. Where possible, a minimum of two health workers per

county per training course were purposively selected for observation. In counties where fewer
than five health workers were in a course, the PTA observed all of the trained health workers.
Table 3 and Appendix 5 list the number of health workers and client interactions observed
during the PTA, by service area.
Table 3: Structured Observations of Health Workers and Client Interactions,
by Service Delivery Area
Service area
Number of health
workers observed
Number of client
interactions observed
HTC 11 14
PMTCT 10 19
Adult ART 33 65
EmONC 27 33
FP/LARC 5 7
TOTAL 86 138
Client Exit Interviews
Following the observations, the clients who had been observed receiving one of the five types of
services were asked to participate in a structured exit interview. One hundred and thirty-eight
(138) clients provided consent. Most clients were female (about 84%), and over two-thirds (69%)
were returning clients. The exit interviews included questions about steps taken to ensure client
confidentiality and clients’ understanding of health workers’ explanations. The client exit
interviews also generated information about the perceived quality of services delivered by
FUNZOKenya-trained health workers and about the service delivery environment.
Data Collection Procedures
The PTA team included FUNZOKenya project staff, representatives from the MOH Human
Resource Department (HRD), regulatory authorities, and CHMT members. Data collection was
carried out in August 2015 by trained research assistants. All data collection tools were
developed through a process of expert/stakeholder review and consensus as well as pretesting.
This process included discussion of how to ask specific questions in Kiswahili. Data collection
staff took part in a three-day training workshop covering collection of quantitative and
qualitative data, including how to probe and write up field notes.
During data collection, the research assistants were divided into five teams to cover the ten
selected counties. In each county, a team of 12 research assistants and two supervisors collected
data over a period of ten working days. Arrangements were made ahead of time with the county
health teams and health facility in-charges to schedule the interviews with supervisors and
county-level key informants.
Informed Consent Procedures
Ethical approval was obtained from the Kenya Medical Research Institute (KEMRI) Ethics Review
Committee and IntraHealth. Respondents were reassured that all information collected would
remain confidential, no individual names would be used for any purpose, and that all
information would be pooled for the report. All participants—key informants, supervisors, health

workers, and clients—provided informed consent before responding to questions or
participating in observations, either via written or fingerprint consent. Where needed,
permission also was sought for audio recording and photography. Interview and survey
participants were allowed to select a time convenient to them.
Quality Assurance
The research team adopted several procedures to improve the quality of data collection. Data
collection supervisors and monitoring and evaluation staff checked data quality immediately
after data collection and held a debriefing session at the end of each day to review challenges
and agree on a way forward. All notes from the qualitative data collection were checked for
completeness and consistency at the end of the day. Upon completion of the structured
interviews with health workers, the data collectors checked the questionnaires for completeness
and accuracy on a daily basis. Data collection supervisors then reviewed the questionnaires and,
where necessary, elicited clarification from the data collectors. After being checked,
questionnaires were allocated individual identification numbers and filed by questionnaire type
in folders that were archived at the project office. Audio recordings and transcripts also were
kept with the project manager.
Data Management and Analysis
Quantitative Data
Semi-structured and quantitative data were entered into SPPS. The data management team
checked for errors by re-entering 5% of questionnaires. Further data cleaning and analyses were
conducted in Stata and Excel. To be conservative, any missing values in the health worker-client
interactions were imputed to “0,” indicating that the task was not completed by the health
worker. The PTA calculated proportions and means. Quantitative results were organized
according to the PTA questions and by category of respondent.
Qualitative Data
All interviews and focus groups were audio-recorded. Immediately after conducting the
interviews and FGDs, summaries were written to supplement notes and audio recordings. The
audio recordings were saved in a computer file and on discs and later transcribed verbatim by
research assistants in the language in which the interviews and discussions were conducted. The
transcripts were then translated into English and entered into Microsoft Word in preparation for
review and coding. A codebook was developed to guide the coding process, which was carried
out using an iterative approach. The coded data, in turn, were entered into NVivo (a software
package for analysis of qualitative data) for content analysis, memo writing, and elaboration of
key concepts. As key ideas and themes emerged, they were grouped and summarized in
accordance with the PTA questions.
Strengths and Limitations
The PTA collected an ambitious amount of data to ascertain the strengths and weaknesses of
the FUNZOKenya-supported training approach and describe the effects of the training on
service delivery. The rich qualitative data collected provide a number of different and important

perspectives on the FUNZOKenya-supported training approach and perceived effects of the
training on services delivered in health facilities.
A major limitation of the PTA was the limited number of structured observations of health
worker-client interactions. For EmONC, because of the difficulty of observing a sufficient number
of specific tasks and services in EmONC service provision, we were not able to conduct robust
analyses of EmONC practices and instead present health workers’ knowledge of EmONC
competencies. Similarly, too few FP/LARC client interactions were observed and too few
FP/LARC-trained health workers were asked knowledge questions to permit analysis and
presentation of robust results. Although the PTA does describe health workers’ practices during
client interactions for HTC, PMTCT, and adult ART services, the small sample sizes mean that
these results should be interpreted with caution and do not necessarily represent the practices
of other health workers trained with FUNZOKenya support in these areas. Another challenge of
this and other similar assessments is the difficulty in disentangling the influence of
FUNZOKenya-supported trainings on observed results from the potential effects of other
training initiatives and health systems support activities.
FINDINGS
In this section, we report our findings related to the training approach and the effects of the in-
service training on the availability and quality of health services provided by health workers
trained with FUNZOKenya support.
Training Approach
Interviewees and focus group participants were invited to comment on the FUNZOKenya
training approach, recapitulated in the box below.
FUNZOKenya Training Approach Highlights
 Identified eight regional training hubs
 Through a competitive process, selected 14 training institutions and practicum sites capable of
delivering in-service training to standards acceptable to Ministry of Health technical departments
 Built training institution capacity for in-service training in key HIV, MNCH, and FP technical areas
 Enhanced collaboration between practitioners, academia-based trainers, CHMTs, and the APHIAPlus
service delivery program.
Overview of Training Approach
Key informants had a number of positive comments about the FUNZOKenya training approach.
Many county-level leaders—such as county AIDS and STI coordinators (CASCOs) and county
health directors (CHDs)—perceived the training institutions as “dynamic” and “ideal” for in-
service training because of their mostly convenient locations, qualified facilitators, and lower
costs. As one CASCO stated, the training institutions “have got the right set-up for conducting

the training.” Although health workers in focus group discussions had fewer comments overall,
they, too, were mostly positive, commenting that FUNZOKenya’s reliance on well-regarded
training institutions made the trainings “credible” and “effective.”
Key informants from partner organizations tended to be more mixed in their opinions. Some
partner organization representatives and most of the county reproductive health coordinators
interviewed reported that the training institutions did not always have suitable facilitators,
sometimes lacked equipment, did not offer adequate practical training, and at times created an
additional layer of bureaucracy. Two APHIAPlus partners openly declared that there was little
meaningful difference between FUNZOKenya’s approach and other trainings. According to one
partner:
At the end of the day, they [FUNZOKenya] use the same facilitators who are from
the Ministry of Health. …The training hubs…were just like…vessels or vehicles to
conduct the same trainings that implementing partners have been conducting all
along.
A second APHIAPlus partner shared similar impressions but noted some benefit in terms of
building the training institutions’ internal capacity—one of FUNZOKenya’s aims. This individual
stated, “The only advantage…maybe is building capacity of those training in institutions.” Further
addressing the topic of capacity building, a training institution representative described it as
“benefit number one” of the institution’s involvement with in-service training:
We as an institution feel involved in…delivery of trainings, whereby it has given us
more power to talk about delivery of in-service trainings, and more health
institutions—private and public—are always looking forward for engagement with
[the] university in terms of this particular service.
Other training institution representatives agreed. One described the training approach as
“perfect” because “it has enriched the training institution and being ahead of what other
organizations have been doing, especially public institutions.” Along the same lines, an
interviewee remarked that the in-service training experience with FUNZOKenya had prompted
the training institution to pursue development of additional short course curricula
“[be]cause…we have innovative things [to offer].” Another university representative concurred,
describing lecturers’ updated skills as a benefit and suggesting that the training experience had
helped bridge “the gap between…the training institutions and the service providers,” helping to
create a more cohesive team that is beneficial for learners.
Assessment participants’ additional comments about the advantages and disadvantages of the
FUNZOKenya training model are presented in the following sections.

Scheduling
Focus group participants and key informants offered various comments and suggestions about
scheduling considerations. Among other points, participants discussed:
 Advance notice: For trainees, the desired amount of advance notice to be able to make
travel arrangements and ensure coverage at work/home prior to attending off-site
training ranged from two days to one month. County-level officials such as CASCOs and
CHDs approved of the collaboration between partners, training institutions, and counties
to schedule trainings, which usually allowed for enough advance preparation time.
 Optimal duration of training sessions: Particularly for ART training, which often covers
PMTCT and adolescent as well as adult ART, focus group participants sometimes felt
shortchanged. Nine participants from four focus groups complained that the trainings
were not long enough to thoroughly cover all the content.
 Time of month and year: Trainees taking part in the focus groups commented on the
least busy and least inconvenient times of month and year to schedule training, with
suggestions ranging from mid-month (when there are no reports to write) to either early
January or mid-year. Key informants acknowledged, however, that there were always
other competing demands and deadlines, and partner organizations noted that
sometimes counties or funders modified their training budget or plans midstream.
 Service interruptions: Trainees, CHDs, and CASCOs all commented on the need to
synchronize planning between county entities and training institutions to minimize
service interruptions as much as possible. Nonetheless, staff shortages invariably result
when providers depart for off-site training. As one CASCO remarked, “Even as much as
we need the training…we cannot be getting people out of facilities for training when
they are supposed to be delivering services.”
Proximity of Training Venues
Some key informants commented on FUNZOKenya’s “50/50” policy, which allowed 50% of
trainees to be residential but required the other 50% to be daytime attendees traveling from
home each day. Several CASCOs expressed frustration with this policy, complaining that
participants who were not offered the residential option had to travel long distances, or that the
distance imposed a constraint on who could be selected to participate: “They said ‘We want
participants who can easily access the training facilities’ and…whoever I wanted to train was not
along the highway so we had to…go [back] to the drawing board.” A representative of one of
the training institutions agreed that the policy sometimes interfered with the flow of training:
“Fifty percent of the participants are supposed to be non-residential…, so we usually
get a lot of challenge with that because for them to go home and come, then that
means some of them will arrive either late or you need to release them earlier for
them to get transport, especially when there are rains.”

Other key informants stated that they would prefer to have input into the selection of training
venues to address the challenges of distance. For example, a county reproductive health
coordinator commented:
…We don’t have an MTC [medical training college], that is why we are being forced
to go up to [a different county]. …If…we can just create somewhere in our resource
center here, instead of going there and using transport.
In the same county, several health workers in focus groups complained about the distance to
the training institution in the neighboring county, noting that “sometimes you were forced to
wake up very early,” but another participant in the same focus group countered that it was
“convenient” because trainees were transported by bus and provided with breakfast. In another
county, a focus group participant stated that the training institutions worked well, but only
because in this instance the training institution facilitators were nearby: “If we were far from
them, we may not have got the information.”
Trainers/Facilitators
Many key informants and focus group participants had positive things to say about the qualified
and “seasoned” trainers at the training institutions. For example, a trainee characterized the
trainers as individuals who are “used to teaching as it is their daily regular profession,”
contrasting these experienced trainers with “a medical person…who does not have the teaching
skills.” Partner organization representatives suggested that the training institutions also were
willing to embrace outside expertise when needed. An informant from a training institution
remarked that the university “delivers quite high training because…majority of our training are
either dealt by professors or senior medical doctors or consultancies in various needs.”
On the other hand, some CASCOs, CHDs, and partners expressed concerns that the training
institution facilitators, being primarily university lecturers, lacked “hands-on” experience relevant
to health workers, particularly for topics such as EmONC. As one CHD put it, “The people they
are training are better well-versed in service delivery than themselves.” A training institution
representative agreed that “If you have an institution that does not do care, and you give them
the mandate to train health care workers, then there is a problem,” but tempered this
agreement with the sentiment that training institutions could do both. Likewise, a trainee
questioned whether the training institutions were familiar enough with the needs of health
workers on the ground. One CASCO elaborated:
Sometimes a [trainer] who is also a health care worker or a [Ministry of Health]
person seems to drive the message home better because they have an experience of
how the health care systems function rather than an outsider. …That is not to mean
that academicians are not as useful, but… where you require practical examples
and practical imparting of the knowledge, sometimes these people who work with
facilities tend to do it better.

Respondents making this point recommended that trainings include a mix of trainers
representing both academia and practice, engaging “hands-on people to be part of the
[training] team.”
Training Materials
Respondents in the focus groups and interviews largely were satisfied with the quality and
appropriateness of training materials. Participants noted that the training manuals:
 Generally derived from approved Ministry of Health curricula
 Were in keeping with established guidelines and standards
 Tended to be readily available at the training institutions.
The only problems noted had to do with challenges acquiring or printing some of the Ministry-
produced materials or lack of funds to purchase mannequins and other classroom aids.
Practical Training
Focus group participants in four of the groups discussed the need for more practice
opportunities as part of in-service training, noting that, without such opportunities, trainees
might return to service and be “confused” about how to perform new clinical duties. Only one
individual described the practice portion of the training as adequate; participants in the other
groups perceived the lack of adequate practical training as a pronounced disadvantage. Trainees
acknowledged that some practice took place on other trainees, but with inadequate time
allotted for trainees with varying levels of experience. Regarding blood draws, for example, a
trainee noted that “Some of us in the lab are used to it, but other people were literally shaking.”
Motivation and Incentives
The topic of health workers’ motivation to attend in-service training attracted numerous
comments from all categories of respondents, who described the challenges of changing the
prevailing culture around per diems. As a training institution representative commented,
trainings can be a “way of motivating [health workers’] performance”—but health workers who
attend trainings “expect to be rewarded.”
Both trainees and key informants noted that hotel-based trainings (a widespread approach in
Kenya) have some undisputed advantages from trainees’ perspective, particularly in terms of per
diems, physical comfort, and the change of pace from normal routines. Commenting on the
“many motivations that make people go for training,” a partner representative listed training
certificates as one factor but also cited allowances, the “opportunity to get away from your
spouse,” and, for some, “never [having] slept in a hotel.” Another partner organization
representative commented, “Some want to just go to the hotel and get the incidental allowances
[and] eat good food in the hotel.”
According to a county health director, the institution-based training environment “is not very
conducive” and is even “demoralizing”; this director laughingly remarked that “we are in service

so we actually need an extra comfort…because…it is just like a classroom, and you are made to
sit there, [and] it becomes a little bit challenging.” In addition, the director commented, “You are
treated like an [undergraduate] student…and even the food being served…, it’s like you are back
in school.” In a focus group, a participant frankly described the training site as “not friendly to
the participants,” because of problems with intermittent water and electricity as well as a dinner
allowance that was “very much low in comparison with other partners who also [do] training.”
A partner organization representative agreed on the need to “reward [trainees] nicely” as an
incentive to perform:
Let the training be conducted in a nice venue. […] [When] you are forced to train
them in a very harsh environment, in rooms without ACs, and you are in a place
that is blazing hot, they feel less motivated. [Mean]while their head in the Ministry
of Health will go to Panafric Hotel in Nairobi for a similar course as a TOT
[training-of-trainers].
Some interviewees discussed resistance to switching over to a less luxurious training model.
According to a representative of one of the training institutions,
When [someone] comes for training, the first thing he thinks of is, “What am I
getting?” though we have really tried to tell them that “We are giving you
information, we are not paying for teaching you.” So that was the initial tug of war
on the…per diem for participants. The next thing was…that we were transferring it
from the hotel-based, there were people who were in charge of that, and they felt
that this cake was being taken, and there was a lot of struggle.
A health worker in a focus group discussion acknowledged that while trainees are “not there to
make money…it is good to induce people with good money.” A CASCO observed that some
trainees are “foolish enough to leave everything [they have learned] there just because they did
not get any money.”
Certification
The topic of training certificates was mentioned in five focus groups and particularly in
Mombasa, where participants called attention to the importance of being certified by a
“recognized institution.” In some of the groups, trainees complained that they could not prove
that they had attended the training because they had never received their certificates. Some
participants noted that other training entities issue certificates online immediately after the
training: “By the time they are finishing the training, they will give you certificates.”
Selection of Trainees
In the focus groups, health workers were asked to discuss how they had been selected to take
part in the FUNZOKenya-supported trainings. Supervisors as well as higher-level decision-
makers also shared their perspectives on the selection process for FUNZOKenya-supported
trainings and in-service training more broadly. A number of focus group participants

(particularly in Mombasa) simply stated that they had been “chosen,” “informed,” or “instructed
to go for training.” Other trainees shared more details. Perhaps because the key informants had
the widest vantage point, they had the most to say about the process used to select trainees.
Selection process. Most key informants described the selection process as a team effort, using
words such as “consultative,” “collaborative,” and “all-inclusive” to describe the “cascading”
involvement of national entities, counties, sub-counties, and facilities. Key informants also
described the interplay between decision-makers such as CASCOs, CHDs, and CHMT members.
A small number of informants mentioned formal training committees as a more recent trend to
guide selection. At the supervisory level, 83% of supervisors reported being involved—to varying
degrees—in selecting trainees to participate in FUNZOKenya-supported trainings. Some
supervisors recommended names to their own supervisors, who then made the final decision,
whereas others provided names of attendees directly to FUNZOKenya.
All three categories of interviewees described the need for a system to assess training needs and
target limited training slots to the “right” or “proper” participants and facilities. Assessment
participants mentioned the need for tools (e.g., training calendars, databases, inventories, log
books, lists) that make it possible for facilities and counties to analyze gaps, identify individuals
who offer the specific services of interest or work in relevant clinics or wards, and differentiate
between individuals who have or have not received prior training in the topic. In some facilities,
these tools already exist; in others, interviewees recommended that they be put in place. A
supervisor commented, “They should take the database, or maybe visit to see the facility and see
the real need and do the assessment of health workers so that you can select those who need
that course rather than duplication.” Another supervisor reported that information of this type
can help identify the health workers most likely to benefit from training, noting that “in the long
run, the patient will also benefit.”
Selection criteria. In the absence of planning tools, a partner organization representative
lamented the potential for training the “wrong” providers: “You…train on long-term family
planning method[s] and you get somebody in maternity.” A training institution representative
formulated the dilemma of selecting suitable participants:
We sometimes [are] provided with participants who are not really directly engaged
in particular service…simply because, at that time, those were the ones available….
Another thing is…, it tends to be your idea when you are asking them to come for
training. So it is not a self-driven initiative. It is more of, “My boss said I attend,” or,
“My head was not there, so I just came.” …I find that as a challenge because you
train this person who will not do anything. It is like waste of money.
Voicing these reservations in a slightly different way, a partner organization
representative commented:

…Most counties don’t have a database of who has received given trainings. […] So
you end up receiving 30 people to train, and you achieve your target, but are those
the right persons you should have trained?
With or without databases, many nonetheless asserted that “the county knows where the gaps
are.” Recognizing that finite training slots exist alongside widespread gaps in skills and services,
most interviewees underscored the importance of “striking a balance” between various selection
considerations and criteria (see Figure 1 below and Table 4 on next page). One CASCO
summarized the balancing act this way:
For the in-service training, the criteria ha[ve] always been, number one, what is the
service need? Number two, has this person been offering the service that is
intended? And number three, what are the gaps they have as individuals?
Figure 1: Key Criteria to Select In-Service Training Participants
One of the most frequently mentioned selection criteria was simply whether staff members had
received training on the designated topic(s) previously. As a provider commented, “…Most of
the providers in the facility had already done it, so I was the only one who was remaining and
needed to update more.” Supervisors confirmed that their selections often were determined by
who had not yet received training on the topic as well as by who they believed would benefit
most. For several facility in-charges who had been trained in areas in which they lacked prior
training, the rewards were immediately evident in the opportunity to return to their facilities to
train other staff and improve their supervisory skills: “…You cannot supervise on things you don’t
know, so by virtue of my position [coordinating reproductive health services] I was selected…to
make me have better supervision skills in the area of reproductive health.”
Trainees, supervisors, and decision-makers discussed the notion that trainees should have some
internal motivation. A reproductive health coordinator commented, for example, that “We want
to identify people who are interested, because you may be trained and you are not interested.”
Similarly, a supervisor linked interest in training to subsequent commitment and performance,

stating, “When selecting, the staffs have to show interest in that particular area of training so
that they can be committed to do the job they are supposed to do.” A CASCO put it this way:
“When you [trainees] go back, you know exactly what is expected of you. You don’t just [go]
fishing somebody anywhere, [and] when they go back [post-training] they become irrelevant.”
Table 4: Training Selection Criteria with Examples
Selection
Criteria
Discussed by:
ExamplesDecision-
makers
Supervisors Trainees
Offer training to
providers or
managers not
previously trained
X X X
 Train new hires
 Respond to transfers or turnover
 Build skills for those who “really need that training”
 Strengthen ability to manage/supervise
Prioritize “key
people”
X X X
Train providers who:
 Provide “hands-on” care
 Have solid performance
 Are capable and qualified but lack training
 Are “able to deliver results”
 Have “interest,” “passion,” “willingness to work”
 “Deserve” the training
Ensure a fair and
equitable
selection process
X X
 Avoid duplication, repeat training, and favoritism
 Equitably distribute opportunities across sub-counties
and cadres
Improve service
delivery in high-
volume or
strategic facilities
X X
 Reduce mismatch between service volume and
number of trained providers
 Ensure skilled staff present where most needed (e.g.,
comprehensive care centers)
 Match work area to training topic so that trainees will
“use that knowledge”
Improve service
delivery in
disadvantaged
facilities
X X
 Train solo providers
 Ensure EmONC skills at facilities far from referral
hospitals
 Balance rural, district, public, private, and FBO facilities
The topic of fairness and unbiased selection elicited almost no commentary from supervisors,
while prompting a small number of remarks in two of the focus groups and somewhat more
discussion in the key informant interviews. Trainees and decision-makers agreed on the
importance of not selecting “the same ones all the time,” but one CASCO acknowledged that
“you discover there are people who have been trained over and over compared to the new
recruits.” A trainee similarly commented, “…In the past, some people repeatedly attended similar
trainings even on those [topics] they have been trained on. This was common due to money and
allowances involved.” These respondents suggested that selection for the FUNZOKenya trainings
had been more impartial than selection for past trainings. Alluding to favoritism and cautioning
on the need to remain “vigilant” against it, a key informant remarked that “if the person in
charge of the health center doesn’t like you, they don’t send you.” A focus group participant
elaborated on this idea:

In the past, many were favored because of how close one was to the bosses or
decision-makers. […] …I would really recommend that [selection for training]
continues that way FUNZO started; however, there is some mischievousness
creeping in again…. I suggest that we restrict ourselves to the database at the
county and sub-county levels to guide us.
From the CHD perspective, equitable distribution of training opportunities has strategic service
delivery implications, ensuring skill development for more than “a few people in the county” and
distribution of trained staff “so that it’s not one sub-county that has more than the other sub-
county.” Partner organizations highlighted strategic considerations such as efficient use of
training resources. A representative of an international organization commented, “There are
times when participants…are chosen and they were trained by another partner for the same
service, and so those are resources that are wasted.” On the topic of appropriate and efficient
targeting, an implementing partner commented that “you don’t want someone who works in the
orthopedic ward to come for a training that is required in the labor ward.”
Supervisors raised one additional topic, namely, the need to expand the overall number of
training slots available. As one supervisor stated, “Many times…you are given the opportunity to
give [training to] one person and maybe there [are]…gaps in more than one person, but you
have been given [only] one slot.”
Post-Training Support
The assessment tried to ascertain from all informants how health workers may have (1) imparted
their acquired knowledge and competencies to others after the FUNZOKenya-supported
training and (2) maintained learned skills and competencies.
Formal mentoring. Over half (51%) of the health workers participating in the assessment’s
structured interviews reported participating in some formal mentorship activities after training.
Half of the health workers who engaged in formal mentoring activities reported passing on their
newly learned competencies to other health workers or sharing skills with support personnel,
students or interns, and/or members of support groups (Table 5). Fewer health workers reported
receiving on-the-job training (23%), routine continuing medical education (CME) (16%), or other
types of formal or partner mentoring (14%) to improve the quality of service provision.
Table 5: Types of Formal Mentorship Activities Reported by Health Workers (n=113) after FUNZOKenya-
Supported Training
Mentorship activity n (%)
Provided training or mentoring to others* on competencies acquired
during FUNZOKenya-supported training 56 (50%)
Received on-the-job training 26 (23%)
Received CME 18 (16%)
Received other formal mentoring 11 (10%)
Partner program (e.g., APHIAPlus) provided formal mentorship/coaching 5 (4%)
*Other health workers, community health workers, support personnel, students or interns, support groups

Tools for Supervision
Checklists
MOH and IMCI tools
Duty rosters
Meetings
Visual or verbal skills
Kardex medical information systems
Spot checks
Charts
Questionnaires
Books
Kenya Quality Model for Health training
Post-training, the majority of supervisors stated that they were concerned with ensuring that
health workers’ knowledge gained during the trainings was shared with colleagues who did not
attend training. Most supervisors strived to arrange for trainees to share new knowledge with
other colleagues via CME meetings held weekly or monthly, depending on the facility. For
example, one supervisor described expectations of trained staff: “After the training they should
come back and give a CME and also be hands-on in that area to implement the services that
they have been trained on.” A few supervisors also reported facilitating on-the-job trainings for
trainees to transfer knowledge to other co-workers: “One, they have to inform us what they have
learnt there and maybe it’s also good to do on-job training for other staff members to ensure
that it is done and then to ensure that what they have been taught to do they are actually
implementing it.”
Routine (integrated) supervision. Supervisors were asked to describe their supervisory
approach as a means of gaining insight into how trainees might be followed up after training.
Most reported using integrated supervision, characterized
by daily supportive supervision and monthly or quarterly
touch-base meetings. Most supervisors reported using
MOH tools for supervision, such as checklists, IMCI tools,
and MOH/NASCOP tools in the areas of HIV, antenatal
care, and immunization, although a wide range of other
tools were also mentioned by a few supervisors (see box).
Supervisors using this integrated approach reported
supervising their staff closely and monitoring staff’s
application of new skills. A supervisor commented that
"This type of supervision is good because you are not
coercing anybody—it is just like mentorship.” Supervisors
also noted that integrated supervision can foster more collaboration with colleagues and
“continuity of knowledge in service delivery” through regular exchanges of information and
updates: “…You are not alone, you belong to a cluster, and also you feel supported…. Two or
three people are more eyes than you and will be able to see the gaps.”
County-level key informants also discussed the advantages of integrated supervision and
suggested that it was more “economical” and “cost-effective” than other forms of supervision
such as targeted supervision, suggesting that “…in terms of cost, it’s much more appropriate,
lower cost, and you are able to achieve more.” Key informants also valued integrative
supervision as it helped them think more “holistically” about health systems issues such as
finance and infrastructure.
Supervisors and key informants admitted that integrated supervision has some disadvantages,
including being time-consuming (thereby adding to supervisors’ already sizeable workloads)
and difficult to fully achieve because supervisors cannot be “everywhere at once.” One key
informant explained, “Number one, we are looking at constrained workforce such that the staff

have so much to do in one sitting.” Another challenge that key informants recognized is that
“supervision is not very deep” since supervisors do not specialize in one focus area.
Training of supervisors. As noted in the section on selection of trainees, a small number of
respondents in the focus groups and interviews commented on the fact that supervisors can
only be effective to the extent that they themselves have been trained and possess the skills that
they are meant to supervise. In addition, according to a focus group participant, facility heads
“should be trained on what it means when we talk about quality measures and management.”
Summary
Table 6 below summarizes the key advantages and disadvantages of the FUNZOKenya-
supported training approach, as perceived by all categories of respondents.
Table 6: Key Advantages and Disadvantages of FUNZOKenya-Supported Training Approach
Advantages
Planning “Consultative” and “collaborative” process to plan trainings and select trainees
Materials Training manuals from approved curricula in keeping with established guidelines and standards
Trainers Qualified and “seasoned” trainers at training institutions who are “used to teaching”
Selection Counties generally adept at “balancing” selection criteria and addressing gaps without favoritism
Sustainability Training institutions not constrained by donor funding cycles and willing to innovate
Disadvantages
Sessions Duration not always sufficient to cover complex material fully (e.g., ART)
Venue Training institution locations not always convenient for attendees traveling from home each day
Trainers University lecturers lack of “hands-on” experience that is relevant to health workers (e.g., EmONC)
Practice Inadequate clinical practice component
Motivation Institution-based environment and lower allowances “demoralizing” (vs. “extra comfort” of hotels)
Recognition Certificates not delivered in timely manner
Reported Effects of Training on Service Delivery
Improved Health Worker Attitudes
A number of health workers, supervisors, and county managers and administrators suggested
that the FUNZOKenya-supported trainings had led to increased confidence and improved
attitudes among trained health workers, often due to the increased knowledge and competency
gained through trainings (Table 7).
Table 7: Trainees’ Perceptions of Increased Confidence and Improved Attitudes Resulting from Training
Improvements Illustrative Quotes
Increased health worker
confidence
 “[In] the areas that I was trained in, I am very confident that [I] am able to perform.”
 “Some of us are from school [a] long time ago, after getting that training at least we
refresh back, at least we know what we are doing.”
 “[I have] improved confidence due to better skills and more acquired knowledge.”
More positive health
worker attitudes
 “The most important is the staff attitude, I realized after the training some of the things
that we were taking for granted had changed, and that was a very positive thing.”

Supervisors suggested that these shifting provider attitudes also influenced the quality of
services:
So if they have HIV-positive client…, at first they were not seeing somebody worth
living. They have changed their attitude. You see they [now] advise them that they
can get a negative baby.
Reported Improvements in Availability and Quality of Services
Nearly all of the interviewed health workers (96%) reported that, post-FUNZOKenya training,
they were providing the services for which they had been trained. For example, over three-
quarters of health workers trained in EmONC with support from FUNZOKenya reported
attending a delivery in the week previous to the survey, and only 4% reported not having
attended a delivery in the past six months. More significantly, all types of respondents perceived
improvements in the availability and quality of services at the facilities of health workers who
had participated in FUNZOKenya-supported trainings, as described in the following sections.
Reported improvements in overall services. Sizeable subgroups of health workers reported
providing additional or improved services after FUNZOKenya-supported training (Table 8),
particularly in the areas of EmONC (53%) and FP/LARC (45%). Some supervisors also reported
that their facilities were now able to offer new and previously unavailable services because
trainees had acquired the necessary skills to offer them. As one supervisor explained, trained
health workers integrate their new knowledge into their practice “and it trickles down to
improved services for the community we serve." The higher-level key informants shared many of
these perceptions about training impact, applauding the trainings for resulting in more up-to-
date and “standardized” care, adherence to clinical guidelines, and a higher quality of care that
eliminates “guesswork.”
Table 8: Health Worker Perceptions of Additional or Improved Services after FUNZOKenya-Supported
Training, by Training Topic
Training topic
Health workers trained in
sample (#)*
Percent (#) reporting additional
or improved services related to
training topic
EmONC 55 53% (n=29)
FP/LARC 44 45% (n=20)
HTC 95 39% (n=37)
ART 92 38% (n=35)
PMTCT 55 31% (n=17)
Newborn care and resuscitation** 60 28% (n=17)
Pediatric ART 32 19% (n=6)
Cervical cancer screening 13 15% (n=2)
*Some health workers were trained in more than one topic and outside the period (Oct. 2013–Sept. 2014) targeted by
the PTA
**Among health workers trained in EmONC or maternal and newborn care

HIV services. In the area of HIV services, supervisors and health workers all reported
improvements in HTC, PMTCT, and ART, citing a stronger ability to prescribe and administer the
right medications and enhanced quality of counseling—with resulting increases in ART uptake
and adherence. Another outcome attributed to the trainings was better integration of HIV
services. As one health worker noted:
After the training, all the departments were involved in HTC because it involved each and
every person working as a health staff; it was an avenue whereby there [was] no missed
opportunity for the client.
Observations of health workers’ practices demonstrated that most health workers who had
received FUNZOKenya-supported training were following HTC and PMTCT testing and
counseling protocols. For example, among health workers who had received training in HTC,
observations of health worker-client interactions showed that the overwhelming majority of
health workers followed infection control practices and HIV pretesting and testing protocols
(Figure 2). Two areas were observed as needing improvement: establishing what prompted the
visit for HIV testing (only 29% completed this task), and reading the test results within 15
minutes (64%).
In observed client interactions with health workers trained in PMTCT with FUNZOKenya support,
the majority of observations noted that health workers who tested expectant mothers during
ANC/PMTCT sessions followed the appropriate procedures during the pretest and post-test
sessions (Figure 3). Health workers were observed explaining to clients how rapid testing is
administered to all clients, and in 79% of health worker-client interactions, health workers
provided information on the importance and benefits of HIV testing and obtained informed
consent. In less than half of client interactions, health workers explained risk reduction and
provided referrals as needed. In the majority of observed interactions with ANC/PMTCT clients
(where all observed results were negative), health workers also provided post-test counseling,
including assessing clients’ readiness to receive results (74%); discussing prevention of future
infections (78%); and exploring and encouraging partner testing (95%) (Figure 3).

Figure 2: Observed HIV Testing Steps by Health Workers Trained in HTC with FUNZOKenya Support
100%
93%
29%
79%
100%
100%
100%
100%
71%
93%
100%
100%
79%
100%
93%
100%
64%
….
Infection control practice observed by the HTC…
MOH HTC lab registers present
…..
Established what prompted visit for HIV testing
Assessed client knowledge of HIV/AIDS
Provided appropriate information on risk…
….
Ensured privacy by closing door/curtain,…
Provided information on benefits of HIV testing
Expressed confidentiality of client
Notified the client their right to refuse to test…
Provided client with ample time to ask questions
Explained what positive, negative and invalid…
Expressed need for consent with HIV testing…
…..
Discussed disclosure of test results and…
Correctly used authorized device…
Healthcare provider wore protective gloves…
Healthcare provider correctly administered…
Provider read the test results 15 minutes after…
Health worker & client safety
HIV testing session
Pretesting session
Obtaining client information
n=14 observations of HIV testing and counseling (HTC) practices among 11 health workers trained
in HTC with FUNZOKenya support

Figure 3: Observed HIV Testing Steps during ANC/PMTCT Visits by Health Workers Trained in PMTCT with
FUNZOKenya Support
Observations of trained health workers’ practices showed variable completion of the steps in the
management of returning ART clients (Figure 4). In the majority of encounters, health workers
assessed the client’s condition, including social support factors (67%), chief complaints and
medical history (58%), and weight and changes in weight (79%). However, less than half (44%) of
clients were screened for TB using the Intensified Case Finding tool. In less than one-third of
client encounters were health workers observed using laboratory assessments to monitor
patients on ART or to assess ART-related toxicities. On the other hand, in approximately two-
thirds of client interactions, trained health workers monitored ART efficacy (60%), managed ART
side effects (67%), and developed or monitored adherence treatment plans (67%). Initiation or
monitoring of cotrimoxazole prophylaxis was completed in less than half (47%) of the observed
encounters.
53%
79%
42%
63%
79%
100%
74%
78%
95%
……
Assessed clients knowledge of HIV/AIDS
Provided information on importance and
benefits of HIV testing
Explained risk reduction and available services
(sexual, MTCT) and provide referrals as needed
Assured client of confidentiality
Obtained informed consent (either verbal or
written) before testing client
Explain how test (rapid testing) is administered?
Assessed clients readiness to receive results
Discussed prevention of future infections
Explored and encouraged partner testing
Pretesting session
Post-testing session (all results negative)
n=19 observations of HIV testing practices with clients in PMTCT/ANC setting among 11 health
workers trained in PMTCT with FUNZOKenya support

Figure 4: Observed Steps Taken in Management of ART Client Visits by Health Workers Trained in ART with
FUNZOKenya Support
EmONC services. Supervisors and health workers credited the training with specific
improvements in EmONC services at their facilities. A supervisor stated, “We used not to fill the
partograph, so the skill has improved and not us alone but also for the other staff in the facility.”
Health workers also mentioned improvements in the active management of third stage of labor,
management of pre-eclampsia and eclampsia, and newborn resuscitation. Some health workers
and supervisors went even further, attributing reductions in neonatal and maternal deaths to
improved skills resulting from the training. As one supervisor remarked, “When encountered
with difficult deliveries, [now] they actually know what to do and [how] to save the life of the
67%
58%
79%
44%
1%
29%
27%
1%
60%
66%
47%
67%
…
Performed social assessment, e.g. family status,
financial status, employment, and social…
Recorded chief complaints and medical history
Weighed patient accurately and calculated
percentage of weight gain/loss
Screened patients for TB using the ICF care tool
….
Performed clinical monitoring of patients on
ART*
Used routine laboratory monitoring to assess ART
associated toxicities*
..
Monitored ART efficacy based on clinical
symptoms, CD4 count, &/or viral load*
Recognized and managed common ART side
effects*
Initiated and/or monitored adherence to Co-
trimoxazole prophylaxis*
Developed/monitored adherence plan with
patient
ART treatment, monitoring, and adherence
Laboratory assessment
n=57 observations of clinical interactions with returning ART clients among 33 health workers trained
in ART with FUNZOKenya support
*Fewer than 57 client encounters included, as the task was not applicable to all clients or done by
other [unobserved] health workers in clinic
Clinical assessment of client's condition

baby.” A health worker echoed this sentiment, “We are saving more babies than before. …After
EmONC, [I] now am to do the resuscitation…and am not waiting for the doctor.” Another
supervisor stated:
It’s really assisting us on how to take care of our patients well. Kilifi County is
known for high mortality…. Previously we used to lose [many] mothers, but for this
year it has really reduced. […] The trainings make them serve patients better.
The PTA was not able to observe enough specific types of obstetrical and newborn client
interactions to assess health workers’ EmONC practices. However, EmONC-trained health
workers were interviewed about knowledge of signs and actions for postpartum hemorrhage
(PPH) (Figure 5) and obstructed labor (Figure 6). The majority correctly identified sub-contracted
uterus (60%), signs of shock (89%), and external bleeding (80%) as signs of PPH, although only
27% identified all PPH signs (Figure 5). Similarly, only 20% of health workers correctly identified
all appropriate PPH management actions, although the majority correctly identified massaging
the uterus (60%); giving oxytocin (87%); starting IV fluids (76%); taking blood for grouping and
matching (58%); and referring to the hospital if bleeding continues (58%).
Figure 5: Knowledge of Signs of and Actions and Interventions for Postpartum Hemorrhage among Health
Workers Trained in EmONC with FUNZOKenya Support
47%
60%
89%
80%
49%
27%
1%
60%
87%
49%
76%
58%
58%
20%
…
Cervical/perineal tears
Sub-contracted uterus
Signs of shock (dizziness, low BP)
Amount of external bleeding
Retained products of conception
Identifed all PPH signs above correctly
…
Massage the uterus /fundus
Give oxytocin IM or IV
Empty the woman’s bladder
Start IV fluids
Take blood for HB & X-matching
Refer to hospital if bleeding continues
Identified all actions above for PPH correctly
Signs of postpartum hemorrhage
n=45 health workers trained in EmONC with FUNZOKenya support surveyed about knowledge
Actions for postpartum hemorrhage

Figure 6: Knowledge of Signs of and Actions for Obstructed Labor among Health Workers Trained in EmONC
with FUNZOKenya Support
Health workers’ knowledge of the signs of obstructed labor was more mixed (Figure 6). The
majority correctly identified no descent of presenting part (82%), Bandl’s ring (51%), maternal
distress (64%), and fetal distress (56%) as signs of obstructed labor, although only 7% identified
all signs. For actions, more than half identified referral (62%) and preparing for Caesarean
40%
29%
20%
82%
29%
36%
51%
64%
56%
7%
1%
22%
38%
40%
53%
44%
62%
9%
…
Cervical dilation rate less than 1 cm per hour
First stage exceeds more than 12 hrs
Second stage is more than 2 hours
No descent of presenting part
Caput
Moulding
Bandl’s ring
Maternal distress
Fetal distress/ Fetal heart rate decreases
Identifed all obstructed labor signs above
correctly
…
Start on 10% dextrose
Empty the bladder
Blood for grouping & cross matching
Prepare for caesarean section
Call the doctor
Refer
Identified all actions above for obstructed labor
correctly
Signs of obstructed labor
n=45 health workers trained in EmONC with FUNZOKenya support surveyed about knowledge
Actions for obstructed labor

section (53%), fewer identified starting intravenous fluids (22%) and emptying the bladder (38%),
and only 9% of health workers correctly identified all actions for obstructed labor.
Family planning services. In the family planning domain, supervisors reported that LARC
uptake had “gone up” because providers who attended training came back and applied their
new insertion skills. As a trained health worker explained, “We were not aware of some of those
methods of family planning, but…now we are able to give the services…so we have improved the
number of our clients seeking…those services.” A supervisor involved in family planning service
delivery reported,
The quality of care has improved so much…. Most of our clients wanted the
reversible methods of family planning, [but]…I worked with the Catholic so they
were not offering those services, [and] I lost experience in between [school and the
job with the Catholic]. So when I went for that training is when I now catch up, and
now we can insert very well implants, IUCDs.
Referrals. Some expanded services were formerly referred out to hospitals. A supervisor
approvingly commented, “It’s a plus for us because [now] you are…doing all the services here.” A
health worker in a focus group discussion also commented on the decrease in referrals: “In my
facility, referrals for clients [were] too high, but…we got the training on how to manage these
cases and in return it has limited our referral cases as we are able to handle them independently
and by ourselves.” Key informants agreed with trainees that in-service training can launch a
positive cycle, whereby providers acquire new skills and offer new services that result in fewer
referrals and better integration of services. According to a county health director:
New officers who…have come from college, they do not have the skill and so…they
keep on referring the patients…elsewhere. The moment they are trained, now they
come back on the ground. They are able to initiate the service and the members of
the community who come to the facility really get that advantage, getting services
nearer them.
Client Experiences
Improved client experiences. Many health workers perceived that the trainings had improved
experiences for their clients. For example, over half (56%) of surveyed health workers reported
that the trainings had led to reduced wait times for clients, and 62% reported improved
confidentiality. Approximately 5% of surveyed trainees noted (without prompting) that the
trainings had improved their counseling of clients. Health workers also raised these positive
benefits for clients in the focus group discussions (Table 9).
Supervisors also commented on aspects of service delivery such as waiting times. However,
supervisors qualified these remarks by pointing out that staff shortages limited the extent to
which in-service training could reduce waiting time. Other supervisors suggested that waiting
times had actually increased because their facilities were offering higher quality of care and, as a
result, attracting a larger client load. One supervisor also described the additional time that

trainees were spending on on-the-job training for their colleagues as another factor causing
waiting times to increase: “You can’t take all people to training, so as they are attending to this
client and they are mentoring their colleague, it takes time.”
Table 9: Trainees’ Perceptions of Improvements in Client Experiences Related to In-Service Training
Improvements Illustrative Quotes
Higher quality provider-
client interactions
“I talk more useful things to him/her compared to the past when we did not have the
facts very much and could spend…time talking about issues which were not…useful.”
Shorter waiting times “The time of waiting has…reduced because we are able to manage the clients who come
to us efficiently and reduce the time wastage because you have the necessary
knowledge to handle the cases.”
Greater client
satisfaction
“They [are] more satisfied because after the training we are giving a range of services.”
A few trainees remarked that providing thorough high-quality care has a down side in terms of
requiring that more time be spent with each client: “Before, you could just examine the client
without taking long, but now you have to go into details on how to take the drugs, so it takes
long.” Another trainee in the same focus group added that when “the contact time between
client and…provider is long,” the result is “more waiting time for others.”
Clients’ satisfaction and reported experiences. Clients rated their clinical experience with
health workers trained with FUNZOKenya support as very positive (Figure 7), with almost 90% of
clients reporting that their experience was either “excellent” or “good.” Clients also were asked
about their perceptions of facility waiting times and about four-fifths (79%) stated that wait
times were adequate, with 21% reporting that wait times were too long.
Figure 7: Clients' Perceptions of Overall Service Delivery Experience (n=138)
Excellent
37%
Good
52%Okay
10%
Bad
1%
n=138 exiting clients seen by health workers trained with FUNZOKenya
support

Asked about various aspects of the health worker-client interaction during their clinic visit, the
overwhelming majority of clients reported having received the services that they were seeking in
interactions characterized by generally good interpersonal communication and respect for
confidentiality (Figure 8). One-fifth of clients reported that they had not been given an
opportunity to ask questions, however.
Figure 8: Clients’ Reported Interactions with Health Workers Trained with FUNZOKenya Support (N=138)
Health System Factors Perceived as Constraining Post-Training Service Delivery
Despite very positive perceptions of the trainings and their effects on service delivery and client
experiences, assessment participants at all levels acknowledged that systemic health systems
challenges often affected health workers’ ability to apply new and/or improved skills to
strengthen service delivery over time. Figure 9 shows the health systems challenges reported by
172 health workers who received training with FUNZOKenya support. Among health workers
citing reasons for not providing new/additional services after FUNZOKenya-supported training
(n=96), a quarter (25%) perceived their current position as not relevant to training. Others cited
various structural factors, including a lack of infrastructure or equipment (16%), supply stockouts
(6%), staff shortages (6%), and non-conducive facility policies (3%). Other reasons were low
demand/client caseload (5%) and lack of confidence in skills (4%). Many trained health workers
in the focus groups also noted that health systems constraints (including lack of opportunity to
practice) sometimes limited the effects of their newly achieved training competencies (Table 10).
96%
94%
94%
94%
78%
98%
Client greeted by provider
Provider asked reason for visit
Provider assured client of confidentiality
of issues discussed
Client understood explanation given by
provider about conditions and needs
Provider gave client opportunity to ask
questions
Client received desired services
n=138 exiting clients seen by health workers trained with FUNZOKenya support

Figure 9: Health Systems Challenges in Delivering Services Reported by Interviewed Health Workers Trained
with FUNZOKenya Support (n=172)
Table 10: Perceived Health Systems Constraints to Applying Training Competencies as Reported in Focus
Group Discussions by Health Workers Trained with FUNZOKenya Support
Illustrative Quotes
 “Knowledge acquired and not used can be forgotten.”
 “When you don’t practice the training, the information just disappear[s].”
 “On [LARC], I am not practicing because I do not have the materials…so I can say I have not been
practicing what I was trained on.”
37%
16%
29%
20%
15%
79%
42%
70%
27%
48%
43%
14%
23%
23%
17%
36%
19%
23%
19%
8%
43%
16%
27%
30%
25%
18%
21%
36%
24%
15%
Inadequate staffing
Heavy workload
Stock-outs
Lack of equipment
Poor/lack of
infrastructure
All interviewed trainees (n=172) HTC (n=33)
PMTCT (n=35) Adult ART (n=64)
EmONC / NN care (n=44) FP/LARC (n=33)
NOTE: Categories of training topics are not mutually exclusive: some trainees were trained on multiple
topics. Values are among interviewed health workers trained with FUNZOKenya support at any time.
*Other health systems contraints cited by health workes included delay in labs/lack of labs (9% in adult
ART and 6% in HTC trained workers) and 42% of HTC-trained health workers reporting issues with the new
HIV (First Response) test.

Illustrative Quotes
 “By the time I came to work in MCH, I had already forgotten everything that I had learnt so I was,
like, I want to be trained again, and even to date, I am still catching up…When you learn
something and you don’t practice, it kind of goes away.“
 “When you take a provider from a facility and take him to where he cannot be able to specialize
on what he had specialized on, it becomes a waste of resources and skills.”
Staff shortages/heavy workloads. The most common health system constraint, cited by over
a third (37%) of health workers interviewed, was inadequate staffing at their facilities, and 16%
cited heavy workloads. Health workers trained in HTC (and assumed to be providing HTC
services) were more likely (79%) to report staff shortages as a major challenge to delivering
services, while health workers trained in FP/LARC were least likely (18%).
Supervisors cited chronic staff shortages, understaffing of health workers possessing the
requisite skills and knowledge, and persistently high client-to-health-worker ratios as the most
pervasive health system challenges, affecting both supervision and other aspects of quality of
care. Many supervisors indicated that the effects of health workforce shortages on supervisors’
high workloads limit their ability to provide necessary support to staff to maintain their skills. As
one supervisor stated, “Sometimes I am so overwhelmed with work that is there, that I don’t
have [time] to do supervising.” Supervisors reported being “disturbed” by the effects of staff
shortages on clients’ ability to receive adequate care, for example, “…You are supposed to sit
with a client or patient for a certain time, but when there is a long queue, you wonder, ‘Do I?
Don’t I?’ and somebody may not know why we are taking ten minutes and there are guys there
who are more sick.”
County managers noted an unforeseen consequence of training staff in new services due to staff
shortages in certain level facilities:
….Now you have introduced a health center where they can go. These people are now
referred to the lower level. They realize these services are available there. So the ones
which were overwhelmed, the numbers go down and probably the waiting time can go
down. But you find for us, …some areas you will find that service providers might still be
overwhelmed because every time, the same service provider who is in the health center
introduce a new service and he is already overwhelmed, he is still challenged, he is still
stretched.
Supervisors described a negative cycle in which staff shortages foster burnout and attrition that
inevitably perpetuates the problem of staff shortages, especially among health workers trained
to deliver specific services. As a supervisor remarked, “You have someone for a short time and
then all of a sudden the person is gone [and] …we have to train them again.”
Shortages of supplies and equipment and lack of infrastructure. The second most common
health system constraint preventing trainees from providing services according to training
competencies, cited by health workers in interviews, was supply stockouts at their facilities

(29%), followed by lack of equipment (20%) and inadequate infrastructure (15%). Once again,
HTC-trained health workers were most likely to cite stockouts as an obstacle to providing
services (70%), followed by health workers trained (and presumably providing services) in
FP/LARC (36%). Reports of lack of equipment and infrastructure were most pronounced among
health workers trained with FUNZOKenya support in EmONC (30% and 25%, respectively) and
HTC (27% and 48%, respectively). Supervisors and key informants also identified the widespread
lack of needed supplies, medications, equipment, and clinic space as factors impeding trained
staff from offering high-quality services.
DISCUSSION
The post-training assessment produced robust data to elucidate the strengths and weaknesses
of the in-service training approach introduced by FUNZOKenya that has trained almost 12,000
health workers over a four-year period. Health workers, supervisors, and county-level and
national-level respondents also described a variety of positive effects on the delivery of HIV,
MNCH, and reproductive health services following the FUNZOKenya-supported in-service
trainings.
The PTA findings suggest that the regional training approach introduced by FUNZOKenya holds
considerable promise. One of the most salient advantages identified by various respondents was
the capacity built at the regional training institutions and the training model’s implied potential
for longer-term sustainability as a result of this enhanced in-country training capacity. This
advantage corresponds to one of the key recommendations put forth by the Health Worker In-
Service Training Improvement Framework developed under the auspices of the USAID ASSIST
Project, which recommends that training programs “use and strengthen local infrastructure and
resources such as trainers, training centers and sites, to improve cost-effectiveness, sustainability
and accessibility of in-service training” (University Research Co. [URC] 2013). The training
approach introduced by FUNZOKenya also responds to other recommendations contained in
the Framework, such as engaging ministries, regulatory and professional bodies, and other key
stakeholders—including significant coordination and consultation at the county level; aligning
in-service training with established national training and HRH plans and priorities; and
supporting continuing professional development of in-country trainers (URC 2013).
Some health workers participating in FUNZOKenya-supported trainings reported dissatisfaction
with the training logistics. Notwithstanding the aim of making in-service training regionally
accessible, the training venues were not convenient for all participants, and some health workers
described the lower allowances and alternative venues (as compared with hotels and conference
centers) as “demoralizing.” Altering trainees’ expectations regarding “motivation” and “rewards”
for participating in in-service training may take time. It also may be worthwhile to explore other
ways to motivate health workers outside of in-service training. Despite citing some
disadvantages, many health workers and supervisors requested that the trainings be expanded.
Respondents expressed interest in including additional topics, offering refresher or updated

training, and accommodating additional staff through more training slots or expanded
schedules.
Respondents’ comments highlighted the importance of providing post-training follow-up,
appropriate supervision, further on-the-job training, and mentorship to ensure that health
workers master and maintain competencies learned in training and are able to put them into
practice (URC 2013; Wasunna et al. 2010). These elements of follow-up also can help health
workers continue to improve the quality of services. The regional training approach launched by
FUNZOKenya needs to include post-training follow-up as an integral component, whether
through enhanced coordination with implementing partners or from the regional hubs
themselves. It is also evident that health systems factors have a substantial influence on the
availability and quality of services, independent of health worker skills or behavior. While most
of the challenges described in this report—such as staff shortages and lack of supplies,
equipment, and infrastructure—are endemic to under-resourced health systems, structured
follow-up after training could potentially provide a forum for discussing and addressing health
systems challenges with health workers, supervisors, health authorities, and implementing
partners.
Evaluations of health worker in-service trainings frequently ask participants to rate their training
experience or track basic indicators, such as the number of people trained (O’Malley et al. 2013).
Few evaluations demonstrate meaningful linkages between in-service training and client, health
systems, or population-level outputs and outcomes, such as enhanced service performance or
greater client satisfaction (Olenja et al. 2009; O’Malley et al. 2013; Penny and Murray 2000),
although these are the desired outcomes of training. The influence of in-service training on
clinical outcomes is far from direct or certain, however (Bluestone et al. 2012). Assessments of
emergency obstetric care training have documented links between in-service training and
provider knowledge, skills, and confidence (Ameh et al. 2012); ability to provide quality
emergency obstetric care (Olenja et al. 2009); and facility performance (Ameh et al. 2012). A
small number of studies have described effects of in-service training on reductions in adverse
neonatal outcomes (Draycott et al. 2006) or neonatal and maternal mortality (Blacklock et al.
2016).
Although the design of the PTA does not allow for meaningful quantitative assessment of the
complex relationship between training and service delivery outcomes, the rich qualitative data
generated by the assessment indicate that training participants, their supervisors, and officials at
the county level and beyond all perceived numerous beneficial effects of the FUNZOKenya-
supported trainings on a variety of intermediate effects, both subtle and straightforward,
including at the client, health worker, and facility levels. Respondents described effects such as
higher-quality interactions with clients; increased health worker confidence and competence;
stronger skills in key service delivery areas such as ART, EmONC, and FP/LARCs; more
integration and availability of services; and correspondingly fewer referrals. In some instances,
respondents also suggested that the trainings had translated into more LARC uptake, better ART
adherence, or a stronger ability to manage obstetric emergencies.

Observations and survey items identified strong knowledge and many strong practices, as well
as some areas for improvement. For example, among health workers who had received
FUNZOKenya-supported training in HTC and PMTCT, the overwhelming majority (often 100%)
were observed to be following infection control practices and HIV pretesting and post-test
counseling protocols. Greater improvement was observed to be needed in only two HTC areas
(establishing the reason for the HTC visit and reading test results within 15 minutes) and two
PMTCT areas (explaining risk reduction and providing needed referrals). Observations of trained
health workers’ management of returning ART clients found that, in the majority of encounters
(58%-79%), health workers assessed the client’s condition, monitored ART efficacy, managed
ART side effects, and developed or monitored adherence treatment plans. Several practices,
however, were observed in less than half of encounters (TB screening, using laboratory
assessments to assess toxicities, and initiating or monitoring prophylaxis). A Zambian study that
examined the effectiveness of a standards-based approach in improving health workers’ ART
performance found ART scores that were roughly comparable to the PTA results, ranging from
58% to 84% in the intervention group and 62% to 70% in the comparison group from baseline
to endline (Kim, Banda et al. 2013). The PMTCT practices observed in the PTA sample were
equally favorable when compared with overall PMTCT scores in the Zambian study (57% to 73%
in the intervention group from baseline to endline) (Kim, Chilila et al. 2013).
Interviews with EmONC-trained health workers identified generally high knowledge levels for
postpartum hemorrhage. The majority of health workers interviewed (60%-89%) correctly
identified three signs of PPH, and roughly three-fifths or more (58%-87%) correctly identified
five PPH management actions. However, only one-fourth (27%) of health workers correctly
identified all PPH signs, and just one-fifth (20%) correctly identified all appropriate PPH
management actions. Overall, the PPH knowledge levels observed in the PTA health worker
sample were either comparable to or higher than levels measured in a 2010 national health
facility survey in Kenya that examined prevention and management of common maternal and
newborn complications (Kagema et al. 2010). The PTA sample’s knowledge of the signs of
obstructed labor was more mixed. Although the majority (51%-82%) correctly identified four
signs of obstructed labor, only 7% correctly identified all signs and only 9% identified all actions
for obstructed labor. PTA respondents’ knowledge levels for obstructed labor were somewhat
lower than levels measured in the 2010 national survey (Kagema et al. 2010), suggesting that
there may be room for further improvement in some areas of EmONC training.
As discussed in the Methods section, the PTA findings must be interpreted with caution. The lack
of a comparison group (temporal or geographic) precludes any conclusive determination about
the relative effects of the FUNZOKenya-supported trainings on the quality and quantity of
services delivered or ultimate impacts on health outcomes. Two supervisors admitted that in the
absence of a formal evaluation, it was not possible to definitively ascertain whether
improvements in client outcomes and service delivery could be attributed to the FUNZOKenya-
supported trainings or to other factors, such as their own performance improvement efforts or
trainings offered by other entities.

RECOMMENDATIONS
A number of recommendations can be gleaned from the PTA findings, including
recommendations for further modifications to the training approach, a stronger practical
training component, and more explicit post-training follow-up and support. Respondents
described a variety of actions that could be taken, encompassing both quantitative and
qualitative aspects of training.
Increase the Quantity, Frequency, and Scope of Training
1. Expand the overall number of training slots available.
2. Schedule trainings more frequently.
3. Provide training on neglected topics such as chronic disease management and supply
chain management.
4. Provide routine refresher training and on-the-job training to accommodate evolving
clinical guidelines.
Use Consistent Criteria and Systems for Trainee Selection and Training Approach
5. Develop and adhere to a consistent system to assess health workers’ training needs and
target training slots to the participants and facilities most in need of training support.
Full implementation of iHRIS Train and its integration with iHRIS Manage at the county
and national levels could facilitate appropriate trainee selection, redeployment, and
retention.
6. Ensure that supervisors receive up-to-date training in the areas in which they are
responsible for supervision.
7. Work toward greater coordination of training approaches across counties by the MOH
and training partners, particularly with regard to allowances, venues, and other logistic
aspects of training.
Ensure that Trainers Have an Adequate Skill Mix
8. Include a mix of trainers representing both academic and clinical settings.
9. Build the “hands-on” training capacity of trainers from the regional training institutions.
Enhance the Practical Training Component
10. Include more practical training, preferably in clinical settings, recognizing that “practicing
on each other isn’t good enough.”

Validate Participation in Training
11. Ensure that completion-of-training certificates are issued without delay, preferably
online, following training.
Provide Meaningful Post-Training Support
12. Strengthen follow-up and provide ongoing mentorship for trainees to ensure solid skills
and performance before skills are passed on to co-workers.
13. Enhance county health departments’ ability to address health system constraints
(including stockouts, lack of equipment, and staff shortages) and their effects on post-
training application of skills learned.
Build Systematic Assessment and Evaluation into In-Service Training
14. At the front end, consider measurement strategies to assess linkages between in-service
training and client, health systems, and population-level outputs and outcomes.
CONCLUSIONS
The findings outlined in this report can inform key stakeholders—including the MOH, medical
training institutions and universities, in-service training beneficiaries, development partners, and
donors—about gaps and needed improvements in training approaches and the health system
supports necessary to ensure health workers can use learned competencies to strengthen
service delivery. The lack of a comparison group precludes any conclusive determination
regarding the relative influence of FUNZOKenya-supported training on the quality and quantity
of services delivered by trainees or its eventual impact on health outcomes. Nonetheless, the
varied evidence collected through the post-training assessment, including a rich assortment of
qualitative data and triangulation of different perspectives, points to potentially meaningful
effects on health worker confidence, skills, and adherence to clinical protocols (particularly for
HIV), with perceived spillover effects that include higher-quality interactions between trained
health workers and clients and greater availability and uptake of key services.

REFERENCES
Ameh, Charles, Adetoro Adegoke, Jan Hofman, Fouzia M. Ismail, Fatuma M. Ahmed, Nynke van
den Broek. 2012. “The impact of emergency obstetric care training in Somaliland, Somalia.”
International Journal of Gynecology and Obstetrics 117, no. 3: 283-287.
Anand S, and T. Bärnighausen. 2004. “Human resources and health outcomes: cross-country
econometric study.” Lancet 364, no. 9445: 1603-1609.
Bailey, Rebecca, Julia Bluestone, Gabrielle O’Malley, Carey McCarthy, Samuel Ngobua, Frances
Petracca, Lois Schaefer, Tana Wuliji. 2013. Improving health worker in-service training to better
deliver quality services. Frontline Health Workers Coalition.
http://frontlinehealthworkers.org/improving-health-worker-in-service-training-to-better-deliver-
quality-services/.
Blacklock, C., D.C. Gonçalves Bradley, S. Mickan, M. Willcox, N. Roberts, A. Bergström, and D.
Mant. 2016. “Impact of contextual factors on the effect of interventions to improve health
worker performance in sub-Saharan Africa: review of randomised clinical trials.” PloS One 11, no.
1: e0145206.
Bluestone, Julia, Peter Johnson, Judith Fullerton, Catherine Carr, Jessica Alderman, and James
BonTempo. 2013. "Effective in-service training design and delivery: evidence from an integrative
literature review." Human Resources for Health 11: 51.
Chen, Lincoln, Timothy Evans, Sudhir Anand, Jo Ivey Boufford, Hilary Brown, Mushtaque
Chowdhury, Marcos Cueto, et al. 2004. "Human resources for health: overcoming the crisis."
Lancet 364, no. 9449: 1984-1990.
Draycott, T., T. Sibanda, L. Owen, V. Akande, C. Winter, S. Reading, and A. Whitelaw. 2006. “Does
training in obstetric emergencies improve neonatal outcome?” British Journal of Obstetrics and
Gynaecology 113: 177-182.
Kagema, Frank, Jim Ricca, Barbara Rawlins, Heather Rosen, Walter Mukhwana, Pamela Lynam,
Nancy Kidula, Mary Gathitu, Chrispin Ndedda, Andolo Miheso, and Rosemary Kamunya. 2010.
Quality of care for prevention and management of common maternal and newborn
complications: Findings from a national health facility survey in Kenya. Are services provided
according to international standards? Kenya: Ministry of Public Health and Sanitation/Ministry of
Medical Services and Maternal and Child Health Integrated Program.
Kawaguchi, A., and R. Mori. In-service training for health professionals to improve care of the
seriously ill newborn or child in low- and middle-income countries. (Last revised: 1 November
2010). The WHO Reproductive Health Library; Geneva: World Health Organization.
http://apps.who.int/rhl/effective_practice_and_organizing_care/CD007071_morir_com/en/.

Kim, Young Mi, Joseph Banda, Webby Kanjipite, Suupriya Sarkar, Eva Bazant, Cyndi Hiner, Maya
Tholandi, Stephanie Reinhardt, Panganani Dalisani Njobvu, Adrienne Kols, and Bruno Benavides.
2013. “Improving performance of Zambia Defence Force antiretroviral therapy providers:
Evaluation of a standards-based approach.” Global Health: Science and Practice 1, no. 2: 213-227.
Kim, Young Mi, Maureen Chilila, Hildah Shasulwe, Joseph Banda, Webby Kanjipite, Supriya
Sarkar, Eva Bazant, Cyndi Hiner, Maya Tholandi, Stephanie Reinhardt, Joyce Chongo Mulilo, and
Adrienne Kols. 2013. “Evaluation of a quality improvement intervention to prevent mother-to-
child transmission of HIV (PMTCT) at Zambia defence force facilities.” BMC Health Services
Research 13: 345.
Ministry of Health. 2014. Post training follow-up assessment of Kenya health workforce trainings
2012. Kenya: Ministry of Health and FUNZOKenya/IntraHealth International.
Ministry of Health. 2011. Report of the performance needs assessment of the Kenya health
training system. Kenya: Ministry of Health and Capacity Kenya/IntraHealth International.
Ministry of Medical Services. 2009. National human resources for health strategic plan 2009–
2012. Nairobi, Kenya: Ministry of Medical Services and Ministry of Public Health and Sanitation.
http://www.nationalplanningcycles.org/sites/default/files/country_docs/Kenya/hrh-strategic-
plan-revised.pdf.
Naidu, A. 2009. “Factors affecting patient satisfaction and healthcare quality.” International
Journal of Health Care Quality Assurance 22, no. 4: 366-381.
O’Malley, Gabrielle, Thomas Perdue, and Frances Petracca. 2013. “A framework for outcome-
level evaluation of in-service training of health care workers.” Human Resources for Health 11:
50.
Olenja, Joyce, Pamela Godia, Josephine Kibaru, and Thaddaeus Egondi. 2009. Influence of
provider training on quality of emergency obstetric care in Kenya. Kenya Working Papers No. 3.
Calverton, Maryland, USA: Macro International Inc.
Penny, Shereen, and Susan F. Murray. 2000. “Training initiatives for essential obstetric care in
developing countries: a ‘state of the art’ review.” Health Policy and Planning 15, no. 4: 386-393.
University Research Co. 2013. A global improvement framework for health worker in-service
training: guidance for improved effectiveness, efficiency and sustainability. Bethesda, MD: USAID
ASSIST Project.
https://www.usaidassist.org/sites/assist/files/inservicetraining_july2013.11x17spreads.pdf.

Wakaba, Mabel, Patrick Mbindyo, Jacob Ochieng, Rose Kiriinya, Jim Todd, Agnes Waudo,
Abdisalan Noor, Chris Rakuom, Martha Rogers, and Mike English. 2014. “The public sector
nursing workforce in Kenya: a county-level analysis.” Human Resources for Health 12:6.
Wasunna, Beatrice, Dejan Zurovac, Jane Bruce, Caroline Jones, Jayne Webster, and Robert W.
Snow. 2010. "Health worker performance in the management of paediatric fevers following in-
service training and exposure to job aids in Kenya." Malaria Journal 9: 261.

APPENDIX 1. SUPERVISOR INTERVIEWS
Forty-one supervisors were interviewed (see table). Interviews took place in all ten counties.
Over half of interviewed supervisors were nurses (n=24); other cadres included physicians,
clinical officers, and lab technologists. Interviewees reported supervising staff that received
FUNZO training on HIV/AIDS (HTC [27%], PMTCT [32%], adult ART [44%], pediatric ART [20%],
and clinical trials units [32%]), EmONC (37%), and FP/LARC (17%). On average, supervisors had
been working in their fields for 14 years.
Characteristics (self-reported)
Number
(Total N=41)
Sex
Female 7
Male 8
Missing 26
Cadre
Physician 1
Clinical officer 7
Nurse 24
Lab technologist 2
Other 1
Missing 6
County
Baringo 2
Homabay 4
Kilifi 6
Kitui 2
Mombasa 5
Nakuru 5
Narok 2
Tharaka-Nithi 6
Trans-Nzoia 4
Vihiga 5
Type of health facility
Public 16
Private 1
FBO 3
NGO 3
Missing 18

APPENDIX 2. KEY INFORMANT INTERVIEWS
Forty-nine key informants were interviewed (see table). Interviews were conducted with county
officials (including health directors, reproductive health coordinators, and CASCOs), national
experts, and representatives of regulatory bodies and training institutions.
Characteristics (self-reported)
Number
(Total N=49)
County
Baringo 4
Homabay 4
Kakamega 2
Kilifi 6
Kitui 1
Mombasa 5
Nairobi 1
Nakuru 3
Narok 3
Tharaka-Nithi 5
Trans-Nzoia 6
Vihiga 3
Missing 6
Job title
County health officer (records, administration) 2
CASCO 8
Reproductive health coordinator 7
County director of health 6
Service delivery manager 5
Technical advisor 2
Clinical position (clinical officer, physician) 4
HIV coordinator 1
FUNZOKenya focal person 3
Deputy chief of party (research and training) 1
Ampath Moi University, MCH/FP technical lead 1
Immunization and nursing database coordinator 1
Training institution manager 3
Regulatory board chairman 1
National office representative 1
Accreditation board member 3

APPENDIX 3. FOCUS GROUP DISCUSSIONS
A total of ten focus group discussions were conducted in eight of the ten counties. The focus
groups involved 79 participants in all.
County
Number of participants
per county
Baringo 8
Homabay 10
Kilifi 9
Kitui 0
Nakuru 7
Narok 8
Mombasa 17
Tharaka-Nithi 10
Trans-Nzoia 0
Vihiga 7
Total 76

APPENDIX 4. HEALTH WORKER INTERVIEWS
Structured interviews were conducted with 222 health workers, of whom two-thirds (67%) were
women (see table). Health workers’ average age was 37.3 years (range=25–57). Nurses/midwifes
represented the largest subgroup of health workers for whom the cadre was reported. The vast
majority were public-sector health workers. Interviews took place in all ten counties.
Characteristics Number of health workers
Sex
Female 149
Male 72
Missing 1
Cadre
Physician 1
Clinical officer 33
Counselor 23
Nurse/Midwife 39
Technologist/Technician 12
Pharmacist/Pharmatech 5
Nutritionist 3
Other 4
Missing 102
County
Baringo 36
Homabay 13
Kilifi 53
Kitui 2
Mombasa 21
Nakuru 32
Narok 36
Tharaka-Nithi 8
Trans-Nzoia 12
Vihiga 9
Type of health facility
Public 194
Private 10
FBO 14
NGO 4
TOTAL 222

APPENDIX 5. OBSERVATIONS OF CLIENT INTERACTIONS
In all, 138 clients were observed in interactions with 86 health workers. Most observed health
workers were nurses (working primarily in adult ART or PMTCT) or nurse-midwives (working in
EmONC). Nearly all observations took place in public-sector facilities ranging from county
referral hospitals to dispensaries. Almost two-fifths (38%) of the facilities were health centers.
HTC
(n=14)
Adult ART
(n=65)
PMTCT
(n=19)
EmONC
(n=33)
FP/LARC
(n=7)
Total
(N=138)
Cadre
Clinical officer — 19 2 — — 21
Counselor 10 — — — — 10
Nurse 2 36 14 — 6 58
Nurse-Midwife — 2 2 33 — 37
Nutritionist — 2 — — — 2
Other 2 6 — — 1 9
Missing — — — — — 1
Sector
Public 12 56 13 30 7 118
Private 2 — — — — 2
FBO — 2 2 1 — 5
Missing — — — — — 13
County
Baringo — 11 10 7 — 28
Homabay — 10 — — — 10
Kilifi 9 — — 6 — 15
Kitui — — — 2 — 2
Mombasa 5 — — — — 5
Nakuru — 10 9 8 — 27
Narok — 4 — 3 6 13
Tharaka-Nithi — 10 — 2 — 12
Trans-Nzoia — 10 — 5 — 15
Vihiga — 10 — — 1 11
Facility Type
County referral hospital 1 10 2 13 3 29
Sub-county hospital — 15 2 8 1 26
Health center 10 27 6 7 2 52
Clinic — — 2 — — 2
Dispensary 3 11 6 3 — 23
Missing — — — — — 6

PTA report 2016March30 FINAL

More Related Content

What's hot (20)

Viewers also liked (6)

Similar to PTA report 2016March30 FINAL (20)

PTA report 2016March30 FINAL