Content uploaded by M. E. Perla
Author content
All content in this area was uploaded by M. E. Perla on Jun 01, 2022
Content may be subject to copyright.
Downloaded from http://journals.lww.com/aidsonline by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/26/2022
Downloadedfromhttp://journals.lww.com/aidsonline by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/26/2022
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
Results of a cluster randomized trial testing
the systems analysis and improvement approach
to increase HIV testing in family
planning clinics
McKenna C. Eastment
a
, George Wanje
b
, Barbra A. Richardson
b,c,d
,
Emily Mwaringa
e
, Kenneth Sherr
b
, Ruanne V. Barnabas
a,b,f
,
Martha Perla
b
, Kishorchandra Mandaliya
b
, Walter Jaoko
g
and R. Scott McClelland
a,b,f,g
Objective: The aim of this study was to test an implementation strategy, the Systems
Analysis and Improvement Approach (SAIA), to increase rates of HIV testing and
counseling (HTC) in family planning clinics in Mombasa, Kenya.
Design: A cluster randomized trial.
Methods: Twenty-four family planning clinics were randomized 1 : 1 to implementing
SAIA versus usual procedures. Study staff implemented monthly SAIA cycles with family
planning clinic staff for 12 months. SAIA has five steps. Step 1 uses a ‘‘cascade analysis’
tool to quantify the number of individuals who complete each step of a process. Step 2
involves sequential process flow mapping to identify modifiable bottlenecks in the
system. Step 3 develops and implements workflow modifications to address bottle-
necks. Step 4 assesses impact of the modification by recalculating the cascade analysis.
Step 5 repeats the cycle. The primary outcome was the proportion of new family
planning clients tested for HIV during the last quarter of the trial.
Results: During the last 3 months of the trial, 85% (740/868) of new family planning
clients were counseled for HIV in intervention clinics compared with 67% (1036/1542)
in control clinics (prevalence rate ratio [PRR] 1.27, 95% confidence interval [CI] 1.15–
1.30). Forty-two percent (364/859) of family planning clients were tested for HIV at
intervention clinics compared with 32% (485/1521) at control clinics (PRR 1.33, 95%
CI 1.16–1.52).
Conclusion: SAIA led to a significant increase in HIV testing in family planning clinics
in Mombasa. Integrating routine HTC into family planning clinics is a promising strategy
to achieve the UNAIDS goal of 95% of people living with HIV being aware of their
status. Copyright ß2021 Wolters Kluwer Health, Inc. All rights reserved.
AIDS 2022, 36:225–235
Keywords: family planning clinics, HIV counseling and testing, implementation
science, systems analysis and improvement approach
a
Department of Medicine,
b
Department of Global Health,
c
Department of Biostatistics, University of Washington, Seattle,
d
Fred
Hutchinson Cancer Research Center, Vaccine and Infectious Disease Division, Seattle Washington, USA,
e
Mombasa County
Department of Health, Mombasa, Kenya,
f
Department of Epidemiology, University of Washington, Seattle, Washington, USA, and
g
University of Nairobi, Medical Microbiology, Nairobi, Kenya.
Correspondence to McKenna C. Eastment, MD, MPH, 325 9th Avenue, Box 359909, Seattle, WA 98104, USA.
Tel: +1 574-210-1120; fax: +1 206-744-3693; e-mail: mceast@uw.edu
Received: 12 October 2020; revised: 28 September 2021; accepted: 5 October 2021.
DOI:10.1097/QAD.0000000000003099
ISSN 0269-9370 Copyright Q2021 Wolters Kluwer Health, Inc. All rights reserved. 225
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
Introduction
Globally, women of childbearing age suffer the highest
burden of HIV infection [1–3]. This is evident in sub-
Saharan Africa where 54% of people with HIV (PWH)
reside, and where 60% of PWH are women [4]. To target
the UNAIDS first 95-95-95 goal, in which 95% of people
know their HIV status, strategies are needed to improve
HIV testing and counseling (HTC) services. In healthcare
systems without a primary care model [5–8], such as
Kenya, family planning clinics offer a unique setting to
integrate HTC for women who do not access other
healthcare services. Since 2008 when HTC data entry
fields were added to the national family planning register,
Kenya’s National AIDS and STD Control Program has
promoted integration of HIV and family planning
services. In Mombasa County, Kenya, 43.9% of married
women were using family planning in the last demo-
graphic health survey (DHS) in 2014 [5]. Contraceptive
use by married women has increased during each DHS
since 2003. Although the DHS only measures contracep-
tive use for married women, it is clear that unmarried
women also access family planning services [9]. Despite
this potential opportunity to increase HTC coverage, a
recent survey of 58 family planning clinics in Mombasa
County found that only 19% of new family planning
clients were counseled for HIV and only half of those
counseled were tested for HIV (10% of total new clients
tested) [10].
Implementation science aims to rigorously address the
‘know-do gap’ [11]. Although there is evidence that an
intervention works (‘the know’) and should be imple-
mented at a population level (‘the do’), how this
intervention should be best implemented presents the
‘gap’. For example, the WHO strongly recommends
integrated care in its Framework on Integrated People-
Centered Health Services [12]. Previous studies have
shown that integrating HTC into family planning services
improves uptake of HIV testing, service quality, and is
inexpensive and feasible [13– 16]. Implementation
science provides a framework for formally testing
strategies to improve integration of HTC with family
planning services [16,17].
Although studies have highlighted the benefits of
integration, there has been concern that integrated
service delivery may be less efficient, more work for
providers, and that clinics lacked adequate staffing and
infrastructure to support integration [18,19]. The Systems
Analysis and Improvement Approach (SAIA), a blended
implementation strategy grounded in systems engineer-
ing, holds potential for addressing these concerns by
systematically identifying and addressing implementation
barriers [20–24]. The objective of this cluster-random-
ized trial was to compare SAIA versus usual procedures for
increasing HTC in family planning clinics in Mombasa
County, Kenya.
Materials and Methods
Study setting
This study was conducted in Mombasa County, Kenya.
The county has an estimated HIV prevalence of 7.4%
[25,26]. The HIV prevalence is higher in women, at
10.5%. In 2018, there were approximately 170 family
planning clinics, including public and private facilities. All
family planning clinics receive commodities at no cost
from the Mombasa County Department of Health
(DOH). Private clinics often charge a convenience fee
for family planning services and HIV testing, which could
impact clients’ acceptance of HTC. All family planning
clinics are expected to follow national and local HTC
guidelines, which suggest that HTC should be performed
for all new family planning clients [27]. In 2008, the
family planning register was updated to capture HTC.
Study design
This was a cluster-randomized trial of 24 family planning
clinics in Mombasa County that were randomized 1 : 1 to
an intervention arm implementing SAIA or a control arm
with usual procedures.
Participants
A survey of HTC performance was conducted in a
random sample of 58 family planning clinics [10]. All 58
clinics were eligible to be enrolled if there were no plans
for clinic closure.
Clinic staff and managers were involved in SAIA’s
implementation and were participants in the trial. Study
staff supported clinic staff by abstracting register data and
supporting idea generation for micro-interventions.
Study staff had no direct contact with family planning
clients. De-identified and aggregated data were collected
from family planning registers.
Baseline data collection
Prestudy baseline HTC data in all new family planning
clients for 2018 were abstracted from family planning
registers. Staff photographed the family planning register,
obscuring client names and contact information. Photo-
graphs were uploaded to a secure encrypted server. Study
staff recorded aggregated data from these images onto
paper CRFs, and data from CRFs were entered into a
REDCap database [28].
Randomization
Randomization was performed by an independent
statistician with no other role in implementing the trial
at the Center for AIDS Research Biometrics Core at the
University of Washington. Restricted randomization was
performed based on clinic size and whether any HTC was
being performed during the initial survey [10]. Of the 58
clinics in this survey, 24 were selected randomly for
inclusion and randomized 1 : 1 to SAIA and usual
226 AIDS 2022, Vol 36 No 2
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
procedures. Because family planning clinic staff partici-
pated in SAIA, blinding could not be performed.
Study interventions and procedures
This study evaluated SAIA, an evidenced-based five-step
multicomponent implementation strategy, focused on
improving care cascades [20– 23]. The first step uses an
Excel-based ‘‘cascade analysis’’ tool to quantify numbers
of individuals who complete each cascade step and
identify priority steps for improvement [24]. Step two
involves sequential process flow mapping to identify
modifiable bottlenecks in the system [21]. Step three
develops and implements workflow modifications
(micro-interventions) to address a bottleneck identified
in step two. Step four assesses the impact of the micro-
intervention and recalculates the CAT. Step five repeats
the cycle.
After family planning clinics were randomized, clinic staff
from intervention facilities were invited for a full-day
SAIA training that launched the trial. This training
included initial facility-level flow mapping, discussion of
each clinics’ CAT, and selection of the first micro-
intervention. The CAT and micro-intervention develop-
ment, testing, and implementation were repeated
monthly for 12 months from December 2018 to
November 2019. Flow mapping was repeated when
clinic flow and processes had substantially changed. Study
staff abstracted family planning register data and
completed the monthly CAT, then reviewed these with
family planning clinic staff. Although study staff
supported family planning clinic staff in developing
micro-interventions, family planning clinic staff were the
implementers of their micro-interventions. Intervention
family planning clinics documented their monthly SAIA
cycles on posters. The posters included the CAT step
targeted, barrier identified, and micro-intervention
proposed. Research staff photographed and transcribed
these implementation plans to create a record of each
SAIA cycle for qualitative analyses.
Clinics randomized to the control arm were informed of
the trial and its objectives. Staff from control clinics were
interviewed to collect data about clinic and staff
characteristics. There were no county-wide initiatives
aimed at increasing HTC in family planning clinics
during the study.
Study staff conducted structured interviews with at least
one staff member from each family planning clinic to
ascertain clinics’ characteristics, including size, location,
whether public or private, whether clients paid for HIV
testing, and how many providers were trained in HTC.
Study staff also asked about availability of private
consultation rooms for HTC to prevent overhearing
conversations (auditory privacy) and seeing client-
counselor interactions (visual privacy). Lack of auditory
or visual privacy may deter clients from receiving HTC.
In addition, clinic staff were asked about clinic managers’
experience, and whether they were familiar with the most
recent Kenyan National HTC Guidelines. Data were
captured on paper case report forms (CRFs) and entered
into a REDCap database [28].
Using the same process for data collection as in the
prestudy baseline period, study staff abstracted data from
the family planning register monthly to complete the
CAT in intervention clinics and quarterly for outcomes
assessment in control clinics.
Several data cleaning steps were undertaken. All CRFs
were reviewed for missing responses. The register image
was reviewed to determine whether data were available. If
so, these were added to the CRF. In addition, CRFs used
to abstract the family planning register data were reviewed
by a second study team member who independently
viewed the digital family planning register, verifying that
the response on the CRF matched the data on the register
image. Any discrepancies were resolved by discussion
between investigators to arrive at consensus about
original register entries. Following data entry, study staff
performed a question-by-question assessment comparing
hard copy CRFs to the digital database to identify and
correct key-in errors.
Outcomes
The primary outcome of this trial was the proportion of
new family planning clients tested for HIV in the trial’s
last 3 months in intervention versus control clinics. Any
client who was documented as known HIV-seropositive
was removed from the denominator for proportion of
new clients tested for HIV. The secondary outcome was
the proportion of new family planning clients counseled
for HIV testing in the trial’s last 3 months in intervention
versus control clinics.
Power and sample size determination
Sample size was calculated assuming clinics would see an
average of 15 new clients per 3-month period, 20% HIV
testing of new family planning clients in control facilities,
and an intervention effect of 50% of clients tested in
intervention facilities. The coefficient of variation of true
proportions between clusters within each group was
estimated at k¼0.5 [29]. With an a¼0.05, a sample of
11 clinics per arm provided more than 80% power to
detect the hypothesized intervention effect. Twenty-four
clinics were included, anticipating that one clinic per arm
might become un-evaluable.
Statistical analysis
Characteristics of clinic staff were presented as counts/
proportions or medians/interquartile ranges. Prevalence
rate ratios were calculated using Poisson regression with a
log link to compare the effect of intervention versus
control conditions on rates of HIV testing during the last
three months of the trial. This primary analysis was
HIV counseling and testing in family planning clinics Eastment et al. 227
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
intended to include adjustment for baseline performance
of HIV testing, as adjusting for baseline covariates can
improve precision and power in randomized trials [30].
However, significant unexpected collinearity was identi-
fied between study arm and baseline performance, so
adjustment for baseline performance was not performed.
Because of this, a secondary analysis was conducted that
tested the difference in differences in HIV testing
between baseline testing and quarter four testing in
intervention and control clinics. Additional exploratory
analyses were performed to determine whether the
intervention effect was different in public versus private
facilities. Results were stratified by public/private status if
the interaction term Pvalue was less than 0.05. A
secondary analysis was conducted that compared the
effect of intervention versus control conditions on rates of
HIV counseling during the last 3 months of the trial, after
adjusting for baseline HIV counseling. Stata version 15
(Stata Statistical Software: Release 15; StataCorp LLC,
College Station, Texas, USA) was used [31].
Qualitative data analysis
Data from implementation plans were summarized and
categorized into types of barriers identified and types of
micro-interventions proposed and executed. Key quota-
tions were selected from these implementation plans to
understand clinics’ micro-interventions.
Ethical considerations
This trial is registered at clinicaltrials.gov (NCT02994355)
and was approved by all ethical and regulatory bodies. The
Mombasa County DOH endorsed this work as a County-
supported public health activity, and clinic managers were
asked to provide assent for their clinics’ participation.
Clinic staff provided verbal assent to participate in
structured interviews. The family planning register data
were fully de-identified, so consent was not required from
individual family planning clients.
Results
Twenty-four clinics were randomized (Fig. 1). Twenty-
three family planning clinic staff participated in structured
interviews. Public and private family planning clinics
across Mombasa County were included (Table 1).
Auditory privacy was available at seven (58%) interven-
tion and eight (73%) control clinics. Visual privacy was
available at eight (67%) intervention and six (55%) control
clinics. Managers were aware of the Kenyan National
HTC Guidelines in six (50%) intervention and nine
(82%) control clinics. The median number of trained
HTC providers was one (interquartile range [IQR] 0–2)
in both intervention and control clinics. HIV counseling
and testing was conducted within the family planning
clinic in three (25%) intervention clinics and two (18%)
control clinics. A convenience fee for HIV testing was
required at two (17%) intervention and three (27%)
control clinics. From baseline data in 2018, counseling
was performed with 29.2% of new family planning clients
in intervention clinics and 46.0% in control clinics. Only
7.3% of new family planning clients were tested for HIV
in intervention clinics compared with 24.7% in
control clinics.
Quantitative results for HIV testing and
counseling
During quarter four, intervention clinics saw a median of
83 (IQR 38–98) new family planning clients and control
clinics saw a median of 79 (IQR 34–253) new clients.
Eighty-five percent (740/868) of new family planning
clients in quarter four were counseled for HIV testing in
intervention clinics compared with 67% (1036/1542) in
control clinics (prevalence rate ratio [PRR] 1.27, 95%
confidence interval [CI] 1.15–1.30). The magnitude of
this effect was larger when adjusted for baseline
performance (adjusted prevalence rate ratio [aPRR]
1.96, 95% CI 1.71–2.24). SAIA’s effect on HIV
counseling was similar in public and private clinics
(interaction term P¼0.7). The rates of HIV counseling
from baseline through each quarter of the trial are
illustrated in Fig. 2.
During quarter four, 42% (364/859) of eligible family
planning clients were tested for HIV at intervention
clinics compared with 32% (485/1521) at control clinics
(PRR 1.33, 95% CI 1.16–1.52). The intervention effect
was different in public versus private clinics (interaction
term P<0.001). SAIA led to more HIV testing in public
intervention versus control clinics (PRR 1.59, 95%CI
1.35–1.23), but not in private intervention versus control
clinics (PRR 0.97, 95% CI 0.77–1.23). The rates of HIV
testing from baseline through each quarter of the trial are
illustrated in Fig. 3.
In the difference in differences secondary analysis, control
clinics tested 24.7% of family planning clients at baseline
and this increased by 5% in the last three months of the
trial. In comparison, intervention clinics tested 7.3% at
baseline and this increased by 35.6% in the last three
months of the trial. This corresponds to a difference in
differences of 32.6% (P<0.05).
There were 7 of 3082 (0.23%) new HIV diagnoses in
intervention clinics and 22 of 5785 (0.38%) new
diagnoses in control clinics.
Qualitative data on micro-interventions
implemented in the intervention clinics
Of the 11 participating intervention family planning
clinics, each participated in 12 monthly SAIA cycles.
Various obstacles were identified by family planning clinic
staff, who proposed a range of micro-interventions to
address these obstacles (Table 2). Barriers to HTC were
grouped into six categories after reviewing clinics’
228 AIDS 2022, Vol 36 No 2
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
HIV counseling and testing in family planning clinics Eastment et al. 229
Fig. 1. Flow diagram of family planning clinics assessed for eligibility, randomized, participated, and included in final intent-to-
treat analyses.
Table 1. Characteristics of 23 family planning clinics and clinic staff included in this trial.
Intervention clinics (n¼12)
N(%) or Median (IQR)
Control clinics (n¼11)
N(%) or Median (IQR)
Public clinic type 6 (50%) 6 (55%)
Urban clinic location 4 (33%) 4 (36%)
Complete auditory privacy in FP clinic 7 (58%) 8 (73%)
Complete visual privacy in FP clinic 8 (67%) 6 (55%)
FP clinic manager aware of most recent National HIV Guidelines 6 (50%) 9 (82%)
Number of FP providers trained in HTC 1 (0– 2) 1 (0– 2)
HIV testing location
In FP clinic 3 (25%) 2 (18%)
In adjacent clinic 8 (67%) 9 (82%)
No HIV testing 1 (8%) 0
Clients required to pay for HIV testing 2 (17%) 3 (27%)
FP, family planning; HTC, HIV testing and counselling; IQR, interquartile range.
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
implementation plans. These included problems with the
family planning register; clinic flow; clinic space; clients
declining HTC; staff availability, knowledge, and
experience; and commodities supply. The most identified
barriers included failure to document HTC in the family
planning register, clients declining HIV testing, staff
turnover, and lack of staff trained to perform HTC. Poor
documentation was addressed early in the study, with 62%
(13/21) of micro-interventions dedicated to documenta-
tion being proposed within the first 3 months of the trial.
230 AIDS 2022, Vol 36 No 2
29.2%
57.7%
72.6%
85.7% 85.3%
46.0%
54.9%
72.3% 66.6%
67.2%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2018 Baseline Quarter 1 Quarter 2 Quarter 3 Quarter 4
Percentage of new FP clients counseled for HIV
tesng
Intervenon Clinics Control Clinics
Fig. 2. Proportion of new family planning clients counseled for HIV testing from baseline through the four quarters of the trial in
intervention and control clinics. Error bars reflect the standard error.
7.3%
33.7%
41.3% 43.7% 42.4%
24.7% 31.1%
31.8%
36.1%
31.9%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2018 Baseline Quarter 1 Quarter 2 Quarter 3 Quarter 4
Percentage of new FP clients tested for HIV
Intervenon clinics Control cl inics
Fig. 3. Proportion of eligible family planning clients tested for HIV from baseline through the four quarters of the trial in
intervention and control clinics. Error bars reflect the standard error.
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
HIV counseling and testing in family planning clinics Eastment et al. 231
Table 2. Types of identified problems, frequency of problems identified, examples of successful micro-interventions, and additional notes and comments from study staff.
Identified problem
Frequency n(%)
of problem identified
and addressed for
one SAIA cycle N¼122 Successful micro-interventions Notes Quotes from FP clinic staff and study staff
Problems with FP register
Poor documentation in family
planning (FP) register
21 (17.2) Improve documentation of counseling and
testing
-Quality control of documentation on a
weekly and monthly basis
These micro-interventions were
proposed early in the study-13/
21 (62%) were proposed in the
first three months of the trial.
‘‘Clinic tested half of eligible clients. QC
[quality control] of register is working.’’
Problems with Clinic Flow
Long wait-times for HIV testing in
FP clinic
2 (1.6) Relocate the HTC services from FP room to
another room
-To use the maternity room for offering
HTC services for all FP clients
‘‘Great improvement in counseling and
testing’’
Long wait-times at voluntary
counseling and testing site (VCT)
where clients are sent for testing
1 (0.8) Integrate HIV testing and counseling (HTC)
into FP services by partitioning the FP
room
Completed in cycle 1 ‘‘Target to increase testing by 40% was
almost realized.’’
Problems with Clinic Space
Lack of room to offer counseling
services due to ongoing
renovations at the hospital
5 (4.1) Allocation for a new room to offer services In this one intervention facility, FP
clinic staff utilized different
available spaces for HTC
because of ongoing hospital
renovations.
‘‘FP nursed liaised with CCC
[Comprehensive Care Clinic] staff to use
the CCC space to offer counseling and
testing services’’
Lack of confidentiality to provide
testing.
1 (0.8) -Find a confidential space to conduct both
HIV testing and FP services
‘‘For the first time, clinic has started testing
clients after creating confidential space
to do testing.’’
Problems with Clients Refusing Counseling or Testing
Clients declining testing in the FP
clinic
33 (27.0) Improve counseling so clients recognize
the benefit of testing
- Improve sensitization through health
talks at clinic
-Involvement of HTC counsellor
-Utilize community health education to
engage community on importance of
testing. Involve community health
volunteers to refer clients to facility.
‘‘All clients counselled and referral system
for testing is in place’’
‘‘Staff sensitized clients on need for
testing. All counselled clients were
tested.’’
‘‘Met target of improving testing by
10%. They counselled all clients’’
Cost implication attached to
testing
2 (1.6) Provide HIV testing by removing
convenience fee for testing in private
facility
‘‘Time taken with clients to counsel them,
and provide testing for free irrespective
of being a private clinic’’
‘‘Facility is offering free HIV testing to all
new FP clients. The testing is being done
at the lab’’
Problems with Clinic Staff availability, knowledge of HTC guidelines, and SAIA
FP staff turnover 22 (18.0) SAIA re-training was offered to all
intervention clinics so that new FP clinic
staff could be trained in SAIA if they
were not part of the initial study launch
training
-Staff trained using standard operating
procedures (SOPs) at the clinic to ensure
no data are missed, and everyone is
aware of the SAIA intervention
Staff turnover was a very common
problem cited
‘‘Weekly and monthly QC. Staff trained on
SAIA’’
‘‘SAIA training was delivered’’
‘‘Great improvement from last cycle.
Making sure all staff understand SAIA
has led to better outcomes’’
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
232 AIDS 2022, Vol 36 No 2
Table 2 (continued )
Identified problem
Frequency n(%)
of problem identified
and addressed for
one SAIA cycle N¼122 Successful micro-interventions Notes Quotes from FP clinic staff and study staff
Lack of staff to do HTC (e.g. HTC
counselor)
15 (12.3) Organize for an HTC counselor to be
stationed in the facility
-Existing staff to ensure they create time
for counseling and testing of clients
-Staff rotation
-Negotiate and liaise with Sub-County
MOH to provide HTS counselor
-Two clinics were finally able to get
HTC counselors into the FP clinic
in cycle 11 after months of trying
‘‘Most improved. The availability of a
counselor meant clients were referred
for testing. Best cycle for the facility
since SAIA began’’
Lack of HIV testing training for
current FP staff
5 (4.1) Train all FP nurses on HTC A proposed micro-intervention that
tried to move HIV testing to the
lab was unsuccessful
‘‘Counseling and testing improved after the
on job training of HTC to all FP staff’’’’
Staff had poor knowledge about
the need for counseling and
testing of clients
4 (3.3) -On the job training for new staff on
counseling and testing
-Sensitizing HCWs to HTC guidelines
-CME to HTC counselor and other staff
‘‘Great success to facility. The micro-
intervention of all staff reminded to offer
HTC and daily QC of record led to
increased rates in counseling and
testing’’
Lack of SAIA training for all
nurses
3 (2.5) Perform SAIA training so that everyone is
aware that all FP clients need to be
counseled and tested
-Introduce SAIA CME to staff
‘‘Nurses trained on SAIA and practical
demonstration in the FP clinic’’
Clients are not referred for testing
(if testing is outside of the FP
clinic)
1 (0.8) Make sure all service providers are aware
of SAIA and refer FP clients for
counseling and testing
‘‘FP staff achieving their target of all clients
counselled and ensuring they are
tested.’’
Problem with Commodities
Lack of HIV testing kits or low
supply of HIV testing kits
7 (5.7) Liaise with SCASCO to provide HIV testing
kits for FP clients.
-Sourcing HIV testing kits from other
locations/clinics
This problem was primarily from
one private clinic (4/7)
‘‘A few kits were sourced from the Sub-
County. They were used to test some of
the counselled clients’’
‘‘Sourced kits from other facilities and
used them to test clients who presented
at the FP clinic’’
No FP products at the moments 1 (0.8) Approach Sub-County MOH on the need
for FP products at the facility
Without FP products, then unlikely
to have FP clients in the clinic to
test for HIV.
‘‘Facility is making sure they have FP
products and encouraging counselling
and testing’’
CME, continuing medical education; FP, family planning; HCWs, healthcare workers; HTC, HIV counseling and testing; MOH, Ministry of Health; QC, quality control; SCASCO, Sub-County AIDS
and STD control officer.
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
Later, micro-interventions targeted other problems and
barriers. Micro-interventions were implemented with
varying degrees of success. Successful micro-interven-
tions included relocating the HTC room to improve
confidentiality, eliminating the convenience fee for HIV
testing in a private clinic, and sourcing HIV test kits from
other clinics when commodities were low. Staff turnover
and scheduled staff rotations were common barriers across
multiple facilities, and these were addressed with a study-
wide SAIA re-training halfway through the study. Two
family planning clinics identified a lack of trained
counselors to perform HTC as a main barrier. One
clinic dedicated three SAIA cycles and another dedicated
five SAIA cycles to getting approval and hiring a
counselor. These interventions ultimately led to the first
family planning clients being tested for HIV in
these clinics.
Discussion
In this cluster randomized trial in Mombasa County,
SAIA significantly increased HTC in new family
planning clients in intervention compared with control
clinics. Initial increases in HTC were likely from
improved register documentation. Later increases were
likely driven by micro-interventions that targeted staffing,
infrastructure, commodity management, and client-
level barriers.
The magnitude of the increase in HIV counseling in
SAIA clinics compared with control clinics was even
larger than the increase in HIV testing. This may be
because family planning providers have more control over
providing counseling than testing. Clients must agree to
testing, including the additional time and potential fee in
private clinics. In addition, clients may need to go to the
laboratory or other testing site if HTC is not offered
directly in the family planning clinic as was the case in
many family planning clinics in this study.
The results from this trial suggest several interventions
that could be useful in eliminating potentials barriers to
HTC in family planning clinics. These include streamlin-
ing testing to minimize the additional time for clinic
visits, performing HTC onsite in the family planning
clinic, and removing any financial costs of HIV testing.
Fees for testing may partially explain SAIA’s minimal
impact on HIV testing in private intervention clinics
compared with public intervention clinics. Anecdotal
evidence from one private clinic supports this hypothesis.
When the clinic waived the HIV testing fee for family
planning clients, their rate of HIV testing increased.
Different micro-interventions implemented during the
SAIA cycles led to improvement in HTC. Other SAIA
trials, including the first SAIA study in the Preventing-
Mother-to-Child Transmission cascade [32], and in the
mental healthcare cascade [33] have demonstrated success
with improving care cascades with micro-interventions.
Recent literature suggests that that continuous monitor-
ing of progress, stakeholder participation, and integration
within existing programs and policies are crucial for
sustainment of implementation strategies in routine
settings [34]. SAIA incorporates these aspects and has
great potential to make sustainable system-level changes.
A 2019 systematic review of research examining
integration of HTC in family planning clinics identified
six studies, and concluded that the evidence generally
supports integration [16]. For example, a nonrandomized
prepost intervention study in Kenya found that provider-
initiated HTC was conducted for 27% of new clients
preintervention compared to 50% following an enhanced
sexual and reproductive health training intervention [35].
In a randomized study from Zambia, there was a
significant increase in HIV testing following enhanced
counseling with referral and an escort to follow-up with
family planning services at 6 weeks. However, despite
promising results at 6 weeks, the impact of this package of
services was not sustained at 6 months [17].
The number of new HIV diagnoses made in this study
was small in proportion to the number of new family
planning clients who were tested. This likely reflects the
changing nature of the HIV epidemic, with undiagnosed
populations becoming harder to find and test [36]. It is
estimated that 90% of Kenyans with HIV know their
status. These individuals contribute to estimates of HIV
prevalence but do not need to be tested [37]. Although
the HIV testing yield in family planning clinics was low,
these clinics remain a valuable setting for HTC because of
their potential for reaching a large segment of the
population of women who are sexually active and at risk
of acquiring HIV.
This study had many strengths. First, the cluster
randomized trial design provides the strongest level of
evidence of SAIA’s efficacy. Second, inclusion of both
public and private facilities in urban and peri-urban
settings supports greater generalizability of these results.
Third, qualitative evaluation of the implementation data
from the micro-interventions helps to highlight how SAIA
was effective at increasing HTC in family planning clinics.
There were also limitations. First, increases in HTC could
not be independently verified through direct observation
of HIV counseling sessions or through increases in
consumption of HIV testing commodities. To mitigate
this limitation, quantitative data were interpreted in the
context of qualitative data from implementation plans.
The barriers identified and the micro-interventions
implemented provide strong evidence that at least some
of the observed increases in HTC were due to actual
changes in practice, rather than simply better record
HIV counseling and testing in family planning clinics Eastment et al. 233
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
keeping for procedures that were already being imple-
mented. Second, due to collinearity, it was not possible to
adjust for baseline differences in HIV testing as planned
for this analysis. To address this problem, a secondary
difference in differences analysis was performed. This
secondary analysis supported the primary results of the
trial. Lastly, this intervention focused on HIV testing, and
did not extend to additional steps in the HIV prevention
and care cascades. However, HTC is a crucial first step in
this cascade.
It is important to understand how SAIA fits into the larger
Kenya HIV Quality Improvement Framework (KHQIF),
which encourages plan-do-study-act (PDSA) cycles for
continuous quality improvement. In essence, SAIA
utilizes PDSA cycles with micro-interventions as ‘Plan
and Do’, the CAT to ‘Study’, and then adopting,
adapting, or abandoning the intervention to ‘Act’ [21,24].
Although we have shown SAIA’s efficacy when delivered
by research staff, the next steps to understanding SAIA’s
sustainability must focus on measuring its effectiveness
when delivered at county-wide scale by County
DOH implementers.
In conclusion, implementation of SAIA by study staff
working with FP clinic staff produced marked improve-
ments in HIV testing in family planning clinics in
Mombasa County. However, testing at entry into family
planning care was not universal even at the end of the
intervention. Given the high proportion of women who
access family planning services, integrating routine HIV
testing for new clients in family planning clinics is a
promising strategy for helping to achieve the UNAIDS
goal of 95% of people living with HIV knowing
their status.
Acknowledgements
M.C.E. has been the primary author on this manuscript
with significant editing by R.S.M. G.W., B.A.R., E.W.,
K.S., R.B., M.P., K.M., W.J., and R.S.M. were involved
in project design, study implementation, and data analysis.
This research was supported by a grant from the National
Institutes of Health/National Institute for Child Health
and Human Development K24-HD88229. M.C.E.
received support from the STD and AIDS Research
and Training Program (T32 AI07140), and through K08-
CA228761. Research reported in this publication was
supported by the University of Washington/Fred Hutch
Center for AIDS Research, an NIH-funded program
under award number AI027757 which is supported by the
following NIH Institutes and Centers: NIAID, NCI,
NIMH, NIDA, NICHD, NHLBI, NIA, NIGMS,
NIDDK. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation
of the manuscript.
Conflicts of interest
R.S.M. received research funding, paid to the University of
Washington from Hologic Corporation, and has received
honoraria for consulting from Lupin Pharmaceuticals.
References
1. UNAIDS. Global Report: UNAIDS report on the Global AIDS
Epidemic 2013. Geneva: UNAIDS; 2013.
2. Bloom DE. 7 billion and counting. Science 2011; 333:562–569.
3. Joint United Nations Programme on HIV/AIDS. UNAIDS data
2017. Geneva, Switzerland: Joint United Nations Programme
on HIV/AIDS; 2017.
4. UNAIDS. Global AIDS Update 2020: siezing the moment -
tackling entrenched inequalities to end epidemics. Geneva,
Switzerland: UNAIDS; 2020.
5. Kenya National Bureau of Statistics, Ministry of Health/Kenya,
National AIDS Control Council/Kenya, Kenya Medical Research
Institute, Population NCf, Development/Kenya. Kenya Demo-
graphic and Health Survey 2014. Rockville, MD: The DHS
Program, ICF International; 2015.
6. Department of Health, Medical Research Council, ORC Macro.
South Africa Demographic and Health Survey 2003. Pretoria,
South Africa: Department of Health/South Africa; 2003.
7. Uganda Bureau of Statistics -UBOS- and ICF International.
Uganda Demographic and Health Survey 2011. Kampala,
Uganda: UBOS and ICF International; 2012.
8. Central Statistical Office/Zambia, Ministry of Health/Zambia,
University of Zambia Teaching Hospital Virology Laboratory,
University of Zambia Department of Population Studies, Tropi-
cal Diseases Research Centre/Zambia, ICF International. Zam-
bia Demographic and Health Survey 2013-14. Rockville,
Maryland, USA: Central Statistical Office/Zambia, Ministry of
Health/Zambia, and ICF International; 2015.
9. FP2020. Kenya: commitment maker since 2012 2020. http://
www.familyplanning2020.org/kenya [Accessed 8 October 2021].
10. Eastment MC, Wanje G, Richardson BA, Nassir F, Mwaringa E,
Barnabas RV, et al.Performance of family planning clinics in
conducting recommended HIV counseling and testing in Mom-
basa County, Kenya: a cross-sectional study. BMC Health Serv
Res 2019; 19:665.
11. Kemp CG, Weiner BJ, Sherr KH, Kupfer LE, Cherutich PK,
Wilson D, et al.Implementation science for integration of
HIV and noncommunicable disease services in sub-Saharan
Africa: a systematic review. AIDS 2018; 32 (Suppl 1):S93–
S105.
12. WHO Department for Service Delivery and Safety WHO. WHO
framework on integrated people-centred health services 2020.
https://www.who.int/servicedeliverysafety/areas/people-
centred-care/en/ [Accessed 8 October 2021].
13. Mansoor LE, Yende-Zuma N, Baxter C, Mngadi KT, Dawood H,
Gengiah TN, et al.Integrated provision of topical preexposure
prophylaxis in routine family planning services in South Africa:
a noninferiority randomized controlled trial. J Int AIDS Soc
2019; 22:e25381.
14. Criniti SM, Aaron E, Hilley A, Wolf S. Integration of routine
rapid HIV screening in an urban family planning clinic. J
Midwifery Womens Health 2011; 56:395–399.
15. Shade SB, Kevany S, Onono M, Ochieng G, Steinfeld RL,
Grossman D, et al.Cost, cost-efficiency and cost-effectiveness
of integrated family planning and HIV services. AIDS 2013; 27
(Suppl 1):S87–S92.
16. Narasimhan M, Yeh PT, Haberlen S, Warren CE, Kennedy CE.
Integration of HIV testing services into family planning services:
a systematic review. Reprod Health 2019; 16 (Suppl 1):61.
17. Hewett PC, Nalubamba M, Bozzani F, Digitale J, Vu L, Yam E,
et al.Randomized evaluation and cost-effectiveness of HIV and
sexual and reproductive health service referral and linkage
models in Zambia. BMC Public Health 2016; 16:785.
234 AIDS 2022, Vol 36 No 2
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
18. Lush L, Walt G, Cleland J, Mayhew S. The role of MCH and
family planning services in HIV/STD control: is integration the
answer? Afr J Reprod Health 2001; 5:29–46.
19. Zotti ME, Pringle J, Stuart G, Boyd WA, Brantley D, de Ravello L.
Integrating HIV prevention in reproductive health settings. J
Public Health Manag Pract 2010; 16:512–520.
20. Rustagi AS, Gimbel S, Nduati R, Cuembelo MeF, Wasserheit
JN, Farquhar C, et al.Implementation and operational re-
search: impact of a Systems Engineering Intervention on
PMTCT Service Delivery in Co
ˆte d’Ivoire, Kenya, Mozambi-
que: a cluster randomized trial. J Acquir Immune Defic Syndr
2016; 72:e68–e76.
21. Wagner AD, Crocker J, Liu S, Cherutich P, Gimbel S, Fernandes
Q, et al.Making smarter decisions faster: systems engineering
to improve the global public health response to HIV. Curr HIV/
AIDS Rep 2019; 16:279–291.
22. Sherr K, Gimbel S, Rustagi A, Nduati R, Cuembelo F, Farquhar
C, et al.Systems analysis and improvement to optimize pMTCT
(SAIA): a cluster randomized trial. Implement Sci 2014; 9:55.
23. Gimbel S, Voss J, Mercer MA, Zierler B, Gloyd S, Coutinho Mde
J, et al.The prevention of mother-to-child transmission of HIV
cascade analysis tool: supporting health managers to improve
facility-level service delivery. BMC Res Notes 2014; 7:743.
24. Wagner AD, Gimbel S,
Asbj€
ornsd
ottir KH, Cherutich P, Cou-
tinho J, Crocker J, et al.Cascade analysis: an adaptable im-
plementation strategy across HIV and Non-HIV delivery
platforms. J Acquir Immune Defic Syndr 2019; 82 (Suppl
3):S322–S331.
25. National AIDS Control Council. Mombasa County HIV & AIDS
Strategic Plan 2016-2020. 2016.
26. National AIDS and STI Control Programme MoHK. Kenya HIV
Estimates 2015. October 2016.
27. National AIDS and STI Control Programme MoHK. The Kenya
HIV Testing Services Guidelines. 2015.
28. Paul A, Harris RT, Robert T, Jonathon P, Nathaniel G, Jose GC.
Research electronic data capture (REDCap): a metadata-dri-
ven methodology and workflow process for providing transla-
tional research informatics support. J Biomed Inform 2009;
42:377–381.
29. Hayes RJ, Bennett S. Simple sample size calculation for cluster-
randomized trials. Int J Epidemiol 1999; 28:319–326.
30. Benkeser D, Dı´az I, Luedtke A, Segal J, Scharfstein D, Rosen-
blum M. Improving precision and power in randomized trials
for COVID-19 treatments using covariate adjustment, for
binary, ordinal, and time-to-event outcomes. Biometrics
2020. doi: 10.1111/biom.13377 [Epub ahead of print].
31. StataCorp. Stata. 15 ed. College Station, TX: Stata Corp LLC;
2018.
32. Rustagi AS, Gimbel S, Nduati R, Cuembelo Mde F, Wasserheit
JN, Farquhar C, et al.Implementation and operational re-
search: impact of a Systems Engineering Intervention on
PMTCT Service Delivery in Cote d’Ivoire, Kenya, Mozambi-
que: a cluster randomized trial. J Acquir Immune Defic Syndr
2016; 72:e68–e76.
33. Fabian KE, Muanido A, Cumbe VFJ, Manaca N, Hicks L, Weiner
BJ, et al.Optimizing treatment cascades for mental healthcare
in Mozambique: preliminary effectiveness of the Systems Ana-
lysis and Improvement Approach for Mental Health (SAIA-
MH). Health Policy Plan 2021; 35:1354–1363.
34. Lennox L, Maher L, Reed J. Navigating the sustainability land-
scape: a systematic review of sustainability approaches in
healthcare. Implement Sci 2018; 13:27.
35. Liambila W, Askew I, Mwangi J, Ayisi R, Kibaru J, Mullick S.
Feasibility and effectiveness of integrating provider-initiated
testing and counselling within family planning services in
Kenya. AIDS 2009; 23 (Suppl 1):S115–S121.
36. De Cock KM, Barker JL, Baggaley R, El Sadr WM. Where are the
positives? HIV testing in sub-Saharan Africa in the era of test
and treat. AIDS 2019; 33:349–352.
37. AVERT.org. HIV and AIDS in Kenya 2020. https://www.aver-
t.org/professionals/hiv-around-world/sub-saharan-africa/kenya.
HIV counseling and testing in family planning clinics Eastment et al. 235
