Study design
The study employed a descriptive cross-sectional design with the aim of finding out the major mediums used in communicating by taking views of a section of users of the communication system at the hospital (health workers and patients). This helped to obtain an overall picture of the communication flow during seeking and provision of health care within the study period.
Setting
The study was conducted at the Komfo Anokye Teaching Hospital (KATH), Kumasi, Ghana on the 30th October to 27th November 2018. The hospital is located in the vibrant and culturally rich city of Kumasi, the regional capital of Ashanti, with a population of about 4.7 million (2010 population census). It is the second largest Teaching Hospital in the country, and the only tertiary institution in the Ashanti Region. It serves as a major referral centre in the northern sector of Ghana. The hospital has (12) clinical directorates, and (2) non- clinical directorates namely surgery, Obstetrics and Gynaecology, Child Health, Medicine, Polyclinic, Diagnostics, Emergency Medicine, Traumatology and Orthopaedics, Oncology, Anaesthesia and Intensive care, EENT, Oral health, Domestics and Technical Services. The hospital has a staff population of 3,909 who fall under these categories, Doctors (9.4%), Top Management (0.2%), Nurses and Midwives (42.2%), Certified Registered Anaesthetist (1.3%), Pharmacist and Pharmacy technicians (3.8%), Administration and Finance (6.6%), Clinical support (10.9%) and Allied Health (5.6%) (KATH Annual Report, 2013).
Participants
Included in the study were health workers such as doctors, nurses, pharmacists, Allied health staff, clinical support staff and directorates managers, who have worked at KATH for at least five years. These categories of staff were included because they had direct communication with the patients during diagnosing, treatment, client service satisfaction as well as payment of services rendered. Staffs who were neither involved in direct provision of health care to clients nor involved in the daily administrative activities of the clinical directorates in relation to patient care were excluded from the study. Again, another set of participants who were included in the study were patients present in the outpatient department in various directorates. This category of patients was included because they are assumed to be in their conscious state and would give reasonable responses on the communication system at KATH. Patients who did not voluntarily consent to be part of the study and those who were not present at the outpatient department of the clinical directorates during the survey were also excluded from the study.
A total of 652 participants consisting of 304 patients, 303 health workers (medical officers, nurses, pharmacists, allied health staff and clinical support staff) and 45 hospital directorate managers were purposively sampled to participate in the study.
Study Outcome
The outcome of interest was effectiveness of all communication mediums used by patients and health professionals during access and provision of healthcare.
Study Predictors
To really get results for the outcome of interest, the study investigated predictors such as the Communication Mediums as used by patients before coming to the hospital and when in the hospital mediums (e.g. Emails, letters, Text messaging, Telephone calls and direct interaction) used by health workers for internal interactions with themselves and with the patient. Lastly, mediums for channeling instructions or information to the staff and to the general public by the hospital managements about an available services. These predictors were assumed would give the clear picture of the effectiveness of the mediums use for patients-health workers interactions during the provision of care.
Data Sources/ Measurement
Primary data was obtained from participants using a structured survey questionnaire. Information was gathered on the communication mediums available for patients’ and health professionals’ interactions and possible challenges emanating from these mediums in Komfo Anokye Teaching Hospital (KATH).
Three different questionnaires designed by the study were used. Each was used independently on the three different populations (health workers, patients and directorate managers). The reason was that each category has different ways of interacting within the hospital hence the different measuring tools.
Prior to the data collection, the questionnaire was pre-tested at the Kwame Nkrumah University of Science and Technology hospital (KNUST) which is in Kumasi to ensure the validity and reliability of the data to be collected from the study. Responses, opinions, and views generated from the pre-testing exercise were used as a guideline to review and refine the data collection tools to be used in the study.
Bias
Respondents of different categories were gathered at one place at a time during tool administration. They were given oral and written explanations as well as necessary instructions and were allowed to complete questionnaires independently.
All questionnaires were labelled numerically to differentiate the completed questionnaires. The completed questionnaires were retrieved on site. Data extracted from the questionnaires was computed and stored on a hard disk. The research document was encrypted to prevent unauthorized access to the stored information.
Study Size
The sample size was estimated using a statistical formula developed by Sullivan (2012) that had a confidence interval (CI) of 95% (α = 0.05), which was used as the standard to make statistical inferences (Frankfort-Nachmias & Nachmias, 2008). This sample was calculated from the study population of health workers, directorate managers and patients using the following formula for dichotomous outcomes Where z is the value from the standard normal distribution for the CI used (e.g. z = 1.96 for 95%); E is the desired margin of error (i.e., 0.5);; and p is the population proportion which is approximate to 0.5.
- sample size for health workers
n = 0.5 (1–0.5) (1.96/0.5)2 = 384
To estimate for non-respondents and incomplete data, 10% was included to approximate to 422.
Sample size for patients
n = 0.5 (1–0.5) (1.96/0.5)2 = 384
To estimate for non-respondents and incomplete data, 10% was included to approximate to 422.
- The sample size for directorate managers
The sample size was calculated using an estimated number of managers selected from 10 directorates with an estimation of N = 50. All the 50 were purposively selected.
Statistical methods
Data was collected using Open Data Kit (ODK) and were transferred onto an excel spreadsheet. It was cleaned and imported into STATA V.14.0 application software for statistical analysis.
To help answer the research questions on the current internal patient-hospital communication mediums used for communications by both patients and healthcare providers, a descriptive analysis was done to describe the counts on each research question and the percentage of responses to each question. This helped the study to get the general picture on the frequency of responses to each research question.