Page 77 - GHES-1-1

Global Health Econ Sustain Clinician’s attitude toward health extension program

the results of the pre-test. The pre-test results demonstrate (MCH) department of the selected facility, assuming that
a similar percentage of clinician knowledge and attitude; it MCH clinicians were directly leading the HEP program.
was 37.5% for good knowledge and a positive attitude. In The data collector initially consulted with the department
light of this, the sample size was established using a single head to determine the number of providers available at
population proportion formula, with a 95% confidence the time of data collection. For instance, the data collector
interval (CI), a proportion of 37.5%, a marginal error of would speak with the HO directly if a department only
4%, a design effect of 2, and a 10% non-response rate, had one. In cases where there were more than two HOs,
and resulting in a total sample size of 1,239 clinicians. We the data collector listed their name in alphabetical order
used a set of 54-item questions to assess the attitudes and and selected one using the Kish-grid method. The same
knowledge of clinicians. procedure was used for selecting clinicians with other
qualifications (“Respondent Selection Methods,” n.d.).
2.3. Sampling procedure
2.4. Data collection
For this study, the term clinician is defined as clinical
staff with the credentials of nurses (diploma and degree), The survey tool was developed by adapting national HEP
midwives (diploma and degree), HO, integrated surgical guidelines and referring to previous studies within the
and obstetricians (ISO), and medical doctors (general field. The final questionnaire was rearranged and reviewed
practitioner, pediatrician, and gynecologist–obstetricians) by experts in public health, HEP, and social education,
who work in public facilities (HC and hospitals) with at such as psychologists, to ensure face and content validity.
least one or more service year(s) in a public facility. This It was written in English and then translated into Amharic.
survey was done under the umbrella of “HEP assessment,” The questionnaire was divided into four sections:
in which data were collected in two rounds, with this sociodemographic data, clinician experience in HEP,
survey being a part of the second round. In the first round, knowledge, and attitude. To identify the outcome variables,
149 HCs from 64 rural districts were selected randomly for we used 54 items (23 knowledge and 31 attitudes), all of
facility assessment. Urban HEP assessment was included which were positively worded. The response for knowledge
in the second round, and an additional 45 HCs (38 from item questions was marked as zero for “No,” or the incorrect
Addis Ababa & 7 from Dire Dawa) were included in the response, and marked as one for “Yes,” or the correct
survey. For the convenience of data collection, all HCs response. The attitude item questions were graded using
(149 HC from the first round and 45 from round two) were a five-point Likert scale, with 0 representing “strongly
included in this survey. disagree,” 1, 2, 3, and 4 representing “disagree,” “neutral,”
“agree,” and “strongly agree,” respectively.
Random selection was employed to choose hospitals
authorized by the Federal Ministry of Health (MoH, 2010). Health background data collectors and supervisors
for practical training of HEWs. As a result, 11 general with bachelor’s and master’s degrees, possessing extensive
hospitals (GH), 15 PHs, and 11 referral hospitals (RH) were experience in survey administration and data collection,
selected. The number and type of clinicians interviewed were recruited and deployed to gather the data through
per facility were chosen using standard staff deployment face-to-face interviews. Before the main data collection
(Resource & Directorate, 2019). For each HC, the research phase, the survey tool underwent a pre-testing process in
team decided to interview five clinicians, including one three districts that were not part of the selected sample.
After a full day of training, field researchers used the
HO, two nurses (degrees and diplomas), and two midwives open data kit tablet app to collect data during interviews.
(degrees and diplomas). In the case of PHs, the number Once the interviews were completed, the collected data
of clinicians interviewed increased to seven, including one was promptly transferred to the main data storage server
HO, two nurses (degrees and diplomas), two midwives through the Internet. Throughout the data collection
(degrees and diplomas), one general practitioner, and one process, a central data manager closely monitored the
ISO. As for general and referral hospitals, the number of incoming data and promptly alerted the field staff if any
interviewees increased to nine, with the addition of one potential errors were detected.
gynecologist–obstetrician and one pediatrician added to
the staff types mentioned for PH. 2.5. Data analysis
We used simple random sampling to select healthcare Data were downloaded from a central server in CSV format
facilities and systematic random sampling to select and exported to STATA version 14 for further data analysis
clinicians. The Kish method was applied within the (STATA Corporation, College Station, Texas 77845 USA).
facility to select the clinicians who would be interviewed. The data were cleaned and recoded after missing values and
The data collector visited the maternal and child health outliers were checked. To ascertain the item distribution,

Volume 1 Issue 1 (2023) 3 https://doi.org/10.36922/ghes.0887