Research approach and design
This study employed an exploratory-descriptive qualitative (EDQ) approach to study perceptions of tertiary-care nurses on factors that promote and hinder acquisition of diabetes knowledge. This approach falls within the premises of naturalistic inquiry [36]. The EDQ approach aims at examining a phenomenon of interest, gaining insights and informing practice [37]. While some authors argue that this approach claims no particular methodologic roots or theoretical underpinnings [36], others reinforce the assertion that the approach is new [37] yet it provides explorative researchers with a “textured, toned and hued” approach that is almost impossible with either a descriptive or an exploratory design [38]. The approach was considered appropriate for the study because it is best suited for exploring perceptions of participants on a particular phenomenon within healthcare settings [39].
Study setting
The study was conducted at two tertiary hospitals, Kamuzu Central Hospital (KCH) and Queen Elizabeth Central Hospital (QECH). These are two of the largest tertiary hospitals in Malawi offering specialist services at regional level and acting as referral centres for district hospitals within their regions [40]. KCH serves a population of 7,523,340 in central Malawi; QECH serves a population of 7,750,629 in southern Malawi [41]. The hospitals are under the Malawi Ministry of Health and Population (MOHP), and have a wide range of subspecialties, such as medicine, surgery, nephrology, neurology, oncology, urology and obstetrics/gynaecology. These facilities were selected because they provide specialised care for patients with NCDs including diabetes and its complications, in addition to the outpatient diabetes clinics which are run weekly [42]. Therefore, most nurses in these facilities are often directly involved in the provision of diabetes care.
Sampling
The study drew its participants from the population of Malawian qualified nurses working at the two tertiary hospitals. The population consisted of both registered nurses and nurse-midwife technicians working in adult care departments. Nurses in these departments deal with patients with diabetes and, therefore, they were expected to be knowledgeable about various aspects of diabetes and its care. This aligns well with the guidelines for the conduct of EDQ research which encourages researchers to obtain participants who are able to provide information required to address the aims of the study [38]. A total of 26 nurses were purposively selected using the following inclusion criteria: (a) nurses working in adult care departments, such as outpatient departments (OPD), short stay wards, critical care units, adult medical and surgical wards; (b) having a minimum of six months practical nursing experience – a period considered as a minimum for familiarization with diabetes knowledge and its associated factors; (c) those in frontline care roles i.e. responsible for and/or involved in providing care to patients with diabetes; and (d) willing to participate in the study. Purposive sampling was utilised in this study because it enables the selection of individuals based on their particular knowledge of a phenomenon for the purpose of sharing that knowledge [43].
Data collection instrument
Data were collected using a semi-structured focus group discussion (FGD) guide. The guide was developed by the researcher (MN) based on the study objectives (see Supplementary Material). The semi-structured nature of the guide allowed the researcher to formulate a written topic guide, which consists of areas or questions to be covered during each FGD [36]. The guide comprised questions relating to the existing barriers and facilitators in the acquisition of diabetes knowledge among nurses at tertiary-care level. The guide also provided an opportunity for nursing staff to make suggestions on the strategies which can be put in place to enhance diabetes knowledge acquisition among nurses. The semi-structured FGD guide was developed in English because it is a medium of instruction in nursing training. However, translations were made whenever necessary during the discussions to facilitate comprehension.
Pretesting of the data collection tool
The FGD guide was pretested with six purposefully selected postgraduate nursing students from Kamuzu University of Health Sciences (KUHeS) in Malawi. Similar to pretesting in the quantitative research, pretesting in qualitative research involves simulating the FGD process on a small scale to identify practical problems relating to data collection instruments, sessions and methodology [44]. MN facilitated the discussion and a research assistant (RA) took notes, given the loose and free-flowing dialogue in FGD often challenging to direct, observe and write at the same time. Observations and corrections were made relating to the structure, flow or order, clarity and appropriateness of the questions as recommended by Polit and Beck (39).
Ethical considerations
The study was approved by the College of Medicine Research and Ethics Committee (COMREC) under KUHeS in Malawi (reference number P.09/22/3766). Before participation, detailed information about the study was provided to all the participants. Subsequently, written informed consent was sought from each study participant. Each participant participated voluntarily with an option to withdraw at any point of the study. Participants were assigned numbers or codes –rather than their names –to ensure anonymity. When data analysis was completed, the focus group transcripts were stored according to the COMREC’s data storage procedures. According to Good Clinical Practice (GCP) Guidelines [45], participants are entitled to reimbursements for expenses incurred as a result of their participation in a study. In this study, the reimbursements included transport costs because the FGD took place in the clinical setting, and compensation for the time lost. Participants were also served refreshments and snacks as appreciation for their availability.
Recruitment process
Recruitment commenced once ethical approval was granted by COMREC. Permission to conduct the study was sought from the Director of each facility prior to submission of the research protocol to COMREC. Posters were pasted on hospital boards and nurses’ stations in the study settings inviting all eligible nurses willing to participate in the study to contact the researcher. Nurses working in the specified departments who met the inclusion criteria were selected. Gatekeepers, such as Unit Matrons were approached to help with recruitment. The researcher also took advantage of nurses’ meetings or morning reports to talk about the research and invite nurses to participate. Moreover, the quantitative phase of the larger study included an option for participants to indicate whether they were willing to participate in the FGD and how they wished to be contacted. Those who expressed willingness to participate were contacted.
Data collection
Data were collected through FGDs by MN from June to September 2023. Two FGDs were conducted with 6–10 participants per group at each tertiary hospital. The date and venue for the FGD were agreed upon by the participants and the researcher. It is recommended that the date and venues for the interviews or discussions be convenient to participants to reduce the researcher-participant power imbalances [36, 37]. Participants were given the freedom to express themselves in a language they were comfortable with, either English or Chichewa, a local language for Malawians. Each discussion lasted between an hour and an hour and-a-half. The description of the FGDs is as follows: firstly, upon entering the discussion room, each participant was welcomed by MN and a RA –a RN who helped with recruitment and note taking at each study site. Subsequently, a wide range of current affairs, such as the weather, were informally discussed with the participants to establish a rapport and make them feel comfortable. Then MN explained the details and the scope of the study to the participants and asked if they needed any clarification on any aspect of the study. Written consent was obtained. Before the commencement of the interview, the facilitator showed the participants the audio recorder and demonstrated the practical mechanics of using it. Consent was also sought from the participants on the use of the recorder. MN started by collecting demographic information of the participants. Literature demonstrates that the collection of quantitative materials, such as demographic information, supports EDQ studies [37]. MN then presented an “ice breaker” to stimulate the participants to start speaking by asking them to react to a scenario in which a nurse had given a wrong dose of insulin. Subsequently, the facilitator introduced questions from the guide and used probes to seek clarification and stimulate discussions. Generally, the FGD explored what the participants perceived as barriers and facilitators in diabetes knowledge acquisition and provided an opportunity for them to suggest strategies for enhancing diabetes knowledge their facilities. The FGD was completed when every group member had finished verbalizing their thoughts or opinions and they had nothing to discuss further. Data saturation was achieved by the end of the third FGD because the participants were free to share their perceptions, thoughts and opinions. However, another FGD was conducted to confirm data saturation. The key feature of data saturation lies with the qualitative researcher’s ability to look for repetition and confirmation of previously collected data [43].
Data analysis
Each recorded FGD was transcribed verbatim by MN. All the discussions were translated into English wherever necessary. A language expert was also consulted to verify the transcription and translation of the recorded discussions. Data were analysed using thematic analysis as proposed by Braun and Clarke (50). Thematic analysis is a method of identifying, analyzing and reporting patterns or themes within qualitative data [46]. Firstly, the transcribed data were read repeatedly by MN to familiarize himself with the data. This process is known as “getting immersed in or dwelling with the data” and is vital for understanding what the data convey [43]. Subsequently, a line-by-line scrutiny or analysis of the transcribed data was conducted to make sense of the data. Colours were used in the sentences to signify potential patterns. Participant quotes were extracted by MN to a matrix. Codes were grouped into thematic categories and were changed when a more exhaustive description emerged. All the authors participated in the data analysis process by discussing and reviewing the identified themes to reach a consensus.
Trustworthiness of data
The study was guided by a framework for ensuring and maintaining rigor in qualitative research proposed by Guba (53) and Guba and Lincoln (54) which has the following components: credibility, dependability, confirmability and transferability. Firstly, the researcher ensured credibility of the findings by member checking and peer debriefing. Member checking was achieved by probing further to seek clarification during the discussions. Peer debriefing was done among the research team following data collection and during data analysis and manuscript preparation. The use of two hospitals also ensured credibility through space triangulation. Secondly, the study ensured dependability by documenting all procedures undertaken in the study including research methods, detailed data collection and analysis. Thirdly, to ensure confirmability, the researcher had an audit trail, which is a recording of all activities of the study including the decisions made over time for others to follow [43]. Lastly, to ensure transferability the study provided rich descriptions of the methodology, participants and their contexts and findings to enable readers or users to consider applying them to other settings or groups.
Study findings
The study had a total of twenty-six (26) participants aged between 22 and 47. There were four males and twenty-two (22) females. The majority of participants (14) were from surgical ward; 8 were from medical ward; and the remaining 4 were from oncology, orthopedic, burns and ear, nose and throat (ENT) departments. Their clinical experience ranged from 7 months to 26 years. Twelve (12) participants were Bachelor’s degree holders; 12 participants were Diploma holders; and 2 participants were Master’s degree holders. The estimated number of patients with diabetes the participants cared for ranged from 0 to at least 15 patients per day.
Data analysis generated the following themes related to the barriers and facilitators in the acquisition of diabetes knowledge among the nurses: (a) individual factors; (b) organizational factors; and (c) innovative suggestions for improving the nurses’ diabetes knowledge (see Fig. 1). The following section describes the themes and their corresponding participant quotes.
Individual factors
Participants in this study identified various factors that hindered their acquisition of diabetes knowledge at an individual level. For instance, some participants mentioned lack of interest to gain new knowledge relating to diabetes. This is illustrated in following excerpt:
“I may also say that it’s because I don’t have the interest to read to increase my diabetes knowledge. I use the knowledge which I acquired in school” (Participant #1 FG2 09).
In addition, one participant explained that they did not feel compelled to gain new knowledge in diabetes because they had not faced challenging situations in their department. This is summarized in the following sentiments:
“But then maybe my emphasis will be at a personal level – self-reflection. #3 alluded that our interest in reading as nurses is minimal. Just to add on that, sometimes what pushes you to go and read is meeting a challenge. For example, for me as a surgical nurse, the thinking is mostly that a surgical patient with diabetes shouldn’t have a worse glycemic status…… and that can’t force me to go and read because I know that when I do what I know the patient doesn’t get worse” (Participant #7 FG1 06).
Participants in this study also explained that they did not have the time to study diabetes. For those who managed to read from the internet, they usually forgot due to pressure of work. The following quote illustrates this:
“Moreover, we don’t have the time to go and read. We might read from Google but you find that you forget due to pressure of work” (Participant #3 FG4 09).
However, the participants also identified various factors that would promote their acquisition of diabetes knowledge at an individual level. For example, one participant emphasized mindset change among nurses considering that there were a lot of coexisting conditions. The following sentiments illustrate this:
“…mindset change. It’s high time we moved away from saying we are specialized in this. We have seen that now there are a lot of coexisting conditions that are bringing patients into the surgical wards… That should be a drive enough for us nurses to search for the information for the conditions that we are dealing with” (Participant #6 FG2 09).
In addition, another participant identified motivation as way of increasing the nurses’ interest in participating in educational strategies to gain new knowledge. This is exemplified in the following excerpt:
“I think there’s also a need to motivate people. Sometimes when you invite people to CPD sessions, they tell you they do not have money for transport. If there can be a certain form of incentive, for example giving them back their days-off people get motivated and say ‘I might need my days-off some day’. Even giving people money would also be a form of motivation…” (Participant #6 FG4 09).
Organizational factors
The study uncovered several factors relating to the organization or system including work environment that impacted on the nurses’ acquisition of diabetes knowledge. For instance, most participants identified work overload as a factor that contributed to their provision of fragmented care and significantly affected their acquisition of diabetes knowledge. This is illustrated in the following participant voices:
“To my thinking, this thing of work overload contributes a lot...the only challenge is the work overload…” (Participant # 4 FG3 09).
“Sometimes you are in a ward and you are taking care of 50 + patients and you have patients with diabetes… You find yourself taking 6–8 hours without checking the blood glucose due to work overload. As such, that also affects the delivery of care. At the end of the shift you’re also too tired to reflect on your practice or check any updates on diabetes” (Participant #3 FG1 06).
Some participants perceived the central hospital set-up as a limitation to their acquisition of knowledge on conditions outside their department. At the same time, they noted that diabetes was becoming a cross-cutting condition:
“I think the first barrier is the system: the hospital and specialisations. So different conditions grouped into one section being managed by people in that particular space. Of course, we are now talking about diabetes as a crosscutting condition; found almost in every department but still more you find a lot of them in the medical ward rather than in the surgical ward” (Participant #6 FG2 09).
Ineffective teamwork among clinical staff members also denied nurses a chance to learn and exchange diabetes knowledge. The collaboration challenges were largely between the nurses and the medical team. This is illustrated in the following excerpts:
“…so you have to run around to get the doctor to do that, so that would be the main the challenge but otherwise the nurse is willing but you have to run up and down for the doctor…. This also denies us a chance to learn more” (Participant #1 FG1 06).
“We leave the other things to doctors but we don’t work as a team. For example, you have a patient who has high blood glucose, you just address the routine problems and leave it there. We hand the patient over to the doctor instead of working together and sharing knowledge. I see this as a common problem for us nurses” (Participant #1 FG4 09).
Participants also identified various factors relating to the hospital administration/management that hindered their acquisition of diabetes knowledge. Some participants believed that the hospital management did not perceive diabetes as a priority. Others pointed out management’s failure to provide trainings and guidelines relating to diabetes. One participant expressed concern over the hospital management’s way of handling nurses who wanted to further their education or specialize in diabetes care:
“If I want to go to school now to do diabetes and I inform the management, they will start “cutting my salary” and even removing me from payroll. I just stay back and say mmm let me continue with what I already have instead of going to school and updating yourself…” (Participant #2 FG1 06).
However, participants identified factors that would promote their acquisition of diabetes knowledge at the organizational level. Some participants thought quality assurance (QA) strategies, such as supervision would be important in practice. For instance, one participant shared that periodic supervisory visits would help to establish the standard of diabetes care. This is illustrated in the following excerpt:
“…there should be periodic supervisory or monitoring visits to see the standard of care nurses are providing to patients with diabetes…” (Participant #4 FG1 06).
Another participant believed that mentorship within the facility could help nurses to gain diabetes knowledge:
“I also just want to add that we can be going to the medical wards –we know that they are experts in such conditions –to be mentored. Maybe once a week, Female Surgical Ward can be sending a nurse there and once this nurse comes back he or she will be able to provide better diabetes care and will be able to mentor others…” (Participant #6 FG2 09).
Effective collaboration between medical doctors and nurses was also perceived as a facilitating factor in practice. This is illustrated in the following excerpt:
“what you need is collaboration between doctors and nurses even if you start with insulin or have the confidence that it is going to help the patient, you can start the treatment and then call the doctor to acknowledge or change the treatment” (Participant #2 FG1 06).
Most participants indicated that the hospital management or organization needed to provide them with in-service trainings considering that diabetes information was constantly changing. This illustrated in the following excerpts:
“To the organization, the in-service trainings that we talked about are very important because things are changing every day... So, the in-service trainings are very important; they can help” (Participant #5 FG4 09).
The majority of participants also highlighted the need to have diabetes management guidelines in all units. They believed that the hospital management was responsible for making the guidelines available in all units:
“As management or policy makers… national diabetes guidelines should be available in all units because we are looking at these patients that they can be found anywhere with different conditions and so there should be that guideline for all the units to have national guidelines for diabetes management” (Participant #4 FG1 06).
Innovative suggestions for improving the nurses’ diabetes knowledge
Participants made innovative suggestions for improving the acquisition of diabetes knowledge on top of the existing ones, such as continuing professional development (CPD) sessions and trainings. One participant cited a simulation-based intervention called “drills” which is done in an emergency setting as one which could also be used to enhance the nurses’ knowledge in diabetes. This is portrayed in the following excerpt:
“I was thinking like in surgical department, sometimes we do what we call drills; drills are like simulation on how staff would respond to an emergency scenario. It helps people who had no interest in the condition to start having interest... So I think if such drills are introduced, they will prompt us to search for knowledge and also we will know our gaps better” (Participant #4 FG1 06).
Another participant indicated that care clubs could be an effective way of organizing nurses to discuss and learn about diabetes or even develop specific guidelines for diabetes management in a group. This is illustrated in the following sentiments:
“Alternatively, you may put us into clubs: some can be in a foot care club; some can be in diet club; others can be in a medication club and so on. We can choose representatives from each group or it can just all nurses in each group to develop our own guidelines –we can do it together or in groups, it’s okay” (Participant #7 FG2 09).