25-08-2017, 09:32 PM
1460140966-document9.pdf (Size: 284.52 KB / Downloads: 5)
Abstract
Background: To increase access to antiretroviral therapy in resource-limited settings, several experts recommend
"task shifting" from doctors to clinical officers, nurses and midwives. This study sought to identify task shifting that has
already occurred and assess the antiretroviral therapy training needs among clinicians to whom tasks have shifted.
Methods: The Infectious Diseases Institute, in collaboration with the Ugandan Ministry of Health, surveyed health
professionals and heads of antiretroviral therapy clinics at a stratified random sample of 44 health facilities accredited to
provide this therapy. A sample of 265 doctors, clinical officers, nurses and midwives reported on tasks they performed,
previous human immunodeficiency virus training, and self-assessment of knowledge of human immunodeficiency virus and
antiretroviral therapy. Heads of the antiretroviral therapy clinics reported on clinic characteristics.
Results: Thirty of 33 doctors (91%), 24 of 40 clinical officers (60%), 16 of 114 nurses (14%) and 13 of 54 midwives (24%)
who worked in accredited antiretroviral therapy clinics reported that they prescribed this therapy (p < 0.001). Sixty-four
percent of the people who prescribed antiretroviral therapy were not doctors. Among professionals who prescribed it,
76% of doctors, 62% of clinical officers, 62% of nurses and 51% of midwives were trained in initiating patients on
antiretroviral therapy (p = 0.457); 73%, 46%, 50% and 23%, respectively, were trained in monitoring patients on the
therapy (p = 0.017). Seven percent of doctors, 42% of clinical officers, 35% of nurses and 77% of midwives assessed that
their overall knowledge of antiretroviral therapy was lower than good (p = 0.001).
Conclusion: Training initiatives should be an integral part of the support for task shifting and ensure that antiretroviral
therapy is used correctly and that toxicity or drug resistance do not reverse accomplishments to date.
Background
Considerable progress continues towards increasing
access to anti-retroviral therapy (ART) in resource-limited
settings. WHO, UNAIDS and UNICEF recently estimated
that 2.12 million people have access to ART in sub-Saharan
Africa, or 30% of people with HIV living there who
need ART [1]. These accomplishments required training of
health professionals, among other efforts to strengthen
health systems. For example, the United States President's
Emergency Plan for AIDS Relief supported the training or
retraining of 219 700 health professionals in ART from
2004 to 2008 [2].
The greatest challenge to increasing access to ART, however,
is the shortage of trained health care professionals
[1,3-5]. Several experts recommend "task shifting" from
doctors to clinical officers and nurses [6-9] or from clinicians
to community health workers [8-10]. According to
WHO, task shifting is the rational redistribution of tasks
among health workforce teams:
"Specific tasks are moved, where appropriate, from
highly qualified health workers to health workers with
shorter training and fewer qualifications in order to
make more efficient use of available human resources
for health." [8]
Gimbel-Sherr et al. demonstrated that expanding the role
of nurses allowed doctors to have more visits with ARTeligible
patients at two clinics in Mozambique [11]. Last
year, they compared ART patients treated by non-physician
clinicians to those treated by doctors, and reported
that the quality of services provided by non-physician clinicians
was equivalent to or slightly better than that of
doctors [12]. Recent articles report on clinical officers
and/or nurses providing ART in Kenya [13,14], Malawi
[15], Rwanda [16] and Zambia [17,18].
In 2008, WHO published global recommendations and
guidelines for task shifting that would promote access to
HIV and other health care services [8]. Recommendation
Four is that countries undertake or update a human
resource analysis on the extent to which task shifting is
already taking place, among other things. Recommendation
Nine is that countries adopt a systematic approach to
harmonized, standardized and competence-based training
that is needs-driven and accredited. The Infectious
Disease Institute (IDI), in collaboration with the Ministry
of Health (MOH) of Uganda, recently conducted a training
needs assessment that addressed both of these recommendations.
Information was collected on the allocation
of ART tasks across health professionals. An audience
analysis [19] provided background on previous training
and self-assessment of HIV and ART knowledge. Uganda was chosen for its well-developed national ART
programme and mature training environment for HIV
care. As of September 2007, an estimated 111 232 people
had access to ART, or 33% of people in need [1]. Uganda
also was a pioneer in training nurses to perform some
tasks of doctors, and lay health workers to perform some
tasks of nurses [20,21]. Health professionals benefited
from several ART training initiatives, including the Drug
Access Initiative [20], and WHO's Integrated Management
of Adult and Adolescent Illness [21], as well as training
from organizations such as the Joint Clinic Research Centre,
HealtheFoundation [22], IDI [20,22,23], Mildmay
International [20,22], Paediatric Infectious Diseases
Clinic at Mulago Hospital, The AIDS Support Organization
(TASO) [22], and Uganda Cares. These organizations
trained a variety of health professionals with courses lasting
from one day to 21 days.
This assessment contributes a method for identifying task
shifting that has occurred in resource-limited settings and
measuring the ART training needs associated with it to the
literature on ART training needs. Previously, the Center
for African Family Studies and Regional AIDS Training
Network conducted an HIV/AIDS training needs assessment
in 12 countries in 2002 that predated scaling-up of
ART (Marc Ahmed Okunnu, personal communication, 10
August 2009). Renggli conducted a situational analysis in
Africa that focused on organizations that provided training
in clinical management of HIV infection, including
ART [20]. Souville et al. reported on knowledge of and
attitudes about ART among physicians in Cote d'Ivoire
[24], and Dohrn et al. reported on knowledge of ART
among midwives in South Africa [25]. The innovative
method presented below can be replicated to inform ART
training programmes in the context of on-going scale-up
and shifting tasks.
Methods
Study design
We surveyed health professionals and heads of ART clinics
at a cross-section sample of clinics that the MOH had
accredited to provide ART. Health professionals reported
on the tasks they performed during a normal work day,
previous HIV training and overall knowledge of HIV and
ART. Knowledge was rated on a six-point scale, where one
was "excellent" and six was "none." The heads of ART clinics
reported on the staff and patients at the HIV and ART
clinic.
Sampling procedure and sample size
We sought a nationally representative sample of accredited
ART clinics in Uganda. The Ugandan health system
divides the country into 11 catchment areas of the
regional referral hospitals. Each area serves several districts.
The national referral hospital in Kampala is the twelfth area; it was excluded from the assessment, because
IDI sought information to guide training programmes for
health professionals outside of Kampala. Using a lottery
method, the following six areas were selected: Arua, Lira,
Masaka, Hoima, Kabale and Mbale.
Using proportionate allocation to size sampling method,
a sample 44 of the 205 accredited facilities as of July 2006
was selected. According to the Ministry of Health, (personal
communication, MOH, National Medical Stores,
2006), 12% of the accredited ART clinics in the six catchment
areas were regional referral hospitals, 35% were district
hospitals, and 54% were health centre IV or III. The
sample included six regional referral hospitals (14%), 16
district hospitals (34%), and 22 health centre IV or III
(52%).
In each catchment area, a random sample of facilities was
selected from a MOH list of accredited ART clinics, stratified
by type of facility. The two strata were: ownership
(government or nongovernmental organization and/or
faith-based organization) and whether or not the facility
was active, i.e. providing ART. Three government district
hospitals were selected randomly from each of the four
catchment areas of the biggest administrative regions and
two from each of the others. One facility that was not
active was selected from each catchment area. At least one
nongovernmental or faith-based facility was selected from
each catchment area, including six hospitals and four
health centres. Two remote facilities were replaced with
proximate ones to stay within the bounds of the study
schedule and budget.
Within health facilities, a convenience sample of health
professionals was selected with the help of the head of the
ART clinic. The inclusion criterion was any person providing
services at the accredited ART clinic who was at the
facility on the day that the study team visited (see below).
We sought to include at least one doctor, clinical officer,
nurse and midwife from each clinic. In Uganda, a doctor
has secondary school education (13 years), five years of
medical school and one year of internship. Clinical officers
are among the non-physician clinicians described in a
recent review [26]; they have a secondary school education,
three years of pre-service training and two years of
internship. There are several types of nurses: all have a secondary
school education; (1) enrolled nurse and enrolled
midwives have one and one-half years of pre-service training;
(2) comprehensive nurses, registered nurses and registered
midwives have three years of pre-service training;
and (3) double-trained nurse-midwives have four and
one-half years of pre-service training
Data collection procedures
Data were collected by means of self-administered questionnaires
for individual health professionals and face-toface
interviews with heads of ART clinics as key informants.
The questionnaires were designed based on examples
from the National Evaluation Center of the United
States AIDS Education and Training Centers. (See http://
aetcnec.ucsf.edu/nec?page=eval-00-00) The questionnaire
for individual health professionals had six sections on (1)
professional background, (2) provision of HIV/AIDS services,
(3) training in HIV/AIDS, (4) barriers to training, (5)
attendance at IDI courses, and (6) IDI's AIDS Treatment
Information Center. The questionnaire for the head of the
ART clinic had similar sections, but only the responses to
questions about staff and patients at the HIV and ART
clinics were used in the analysis.
Early versions were shared with stakeholders representing
HIV training organizations in Uganda in a participatory
process that led to several improvements. Later versions of
the questionnaires were pretested with health professionals
and the head of the ART clinic at Mbuya Reach Out and
a Kampala City Council clinic. The final questionnaire for
health professionals is included as Additional file 1; the
questionnaire for the head of the ART clinic is available
from the authors on request.
Twelve research assistants were trained in data collection
for three days. They were grouped into four teams, each
comprising a social scientist, medical doctor and field
assistant. A team spent one day at each facility and collected
data from an average of 11 facilities during a twoweek
period in July and August 2006.
Data management and analysis
The completed questionnaires were coded and the data
were double-entered in Epi-Info version 6.01 software
(Centers for Disease Control and Prevention, Atlanta GA,
United States of America) to ensure accuracy and integrity
of the data. Descriptive statistics and statistical tests were
conducted with SPSS-PC software, version 11.0 for Windows
(SPSS Inc, Chicago IL, United States of America).
Data analyses were stratified by health profession and chisquare
(χ2) tests were used to assess statistical significance
of differences in proportions (percentages). Where there
were a small number of cases (expected frequency less
than 5); Fisher's exact tests were used.
Human subjects
The study was approved by the IDI training committee,
the MOH and the Institutional Review Board of the Faculty
of Medicine at Makerere University. Respondents provided
oral informed consent.
Results
Characteristics of the sample
Forty-three of the 44 facilities selected were included; a
team was unable to travel to one nongovernmental health
centre that was not active. Thirty-eight of the 43 health
facilities were active and five (one district hospital and
four health centre IVs) were not. As shown in Figure 1, the
regional referral hospitals provided ART to an average of
1727 HIV patients per month, whereas the district hospitals
and health centre IV provided ART to an average of
228 and 78 people, respectively. Regional referral hospitals
reported the highest proportion of HIV patients
receiving ART (45%), while 33% and 17% of HIV patients
received ART at district hospitals and health centre IVs,
respectively.
The sample of health professionals included 265 clinicians:
34 were doctors, 46 clinical officers, 124 nurses and
61 midwives. Sixty percent were female and 58% were
aged 35 years or younger. Table 1 compares the respondents
to the staff that the head of the ART clinics reported
were assigned to the ART clinics. The distribution of
respondents across health professions differed significantly
from the distribution of staff assigned to the ART
clinics. Doctors were underrepresented at all types of facilities;
nurses were underrepresented at regional referral
hospitals and district hospitals and overrepresented at
health centre IVs. No doctors were on the staff of ART clinics
at two district hospitals and two health centres.