The study utilized a pre-and post-test non-randomized quasi-experimental research design consisting of two groups of nurses selected from different hospitals, with both groups receiving the intervention and a comparison made to assess the effectiveness of the intervention. The research aimed to determine the effect of training on nurses' knowledge, attitude, and practice in pain management. The design involved the comparison of participants before and after implementing the training programme. It also involved the collection of baseline data that allowed the researcher to be relatively confident, inferring that post-test differences would occur as a result of the intervention; thus, the design was deemed appropriate for this study.

Nurses were recruited from two government-owned health facilities in Ilorin, Kwara State, Nigeria. The two hospitals are referred to as Hospitals A and B in this study. Hospital A is a secondary healthcare institution providing out-patient and in-patient services. Other services provided included but are not limited to medical, surgical, and dental services, amongst others. While Hospital B is a government-owned secondary healthcare institution that provides both out-patient and in-patient activities, medical and surgical services, with dental services, amongst others.

The target population for the study were nurses working in Hospitals A and B, with a total of 152 nurses, consisting of 104 and 48 from Hospitals A and B, respectively. The sample size was estimated using Slovin’s formula [18]. After factoring in the 10% attrition rate, the total sample size was 121. The samples consisted of 83 and 38 participants from Hospitals A and B, respectively.

The multistage sampling technique with three stages was adopted for the study. In stage one, a simple random sampling technique was used to select two out of three local government areas in the study setting, and one hospital each was selected from the two local government areas, making a total of two hospitals. In stage two, a stratified sampling technique was used to select 83 and 38 nurses from the two selected hospitals using the total number of nurses. The convenience sampling technique was used in the third stage to recruit the required number of participants. The inclusion criteria for the study include being a registered nurse or midwife; they must have practiced for at least a year and worked in the selected hospitals. Registered nurses or midwives with prior pain management training were excluded from the study.

The instrument for data collection was a questionnaire adapted from similar studies [12, 19-21]. The instrument was modified so as to evaluate the content to know if it reflects the concepts/objectives of the study. The content validity was also measured after a thorough literature review, and the questionnaire was also given to experts in the field of pain management for their corrections and improvement. The test-retest approach was used to determine the data's dependability. After the expert review, to ensure its reliability, the tool was completed by 15 nurses. The results were, however, not added to the final data presentation. Also, the reliability coefficient was measured to test for internal consistency. The data collected was analyzed, and Cronbach’s alpha value of 0.7 and above was accepted as reliable. The modified questionnaire consisted of 60 items that cut across the subscales of Nurse Knowledge, Attitude, Survey, and Practice (NKAS) on pain. Sections A and B of the questionnaire consisted of the participants’ demographical characteristics and 25 knowledge questions, based on ‘Yes’ or ‘No’ responses, with a total range of 0–25 points. The scores were converted into percentages, and participants with scores below 50% were categorized as having inadequate knowledge, while those with scores above 50% had adequate knowledge. Section C was on attitudinal disposition with a total of 20 questions on a four-point Likert Scale from strongly agree, which equals 4, to strongly disagree, which is 1, and the total score ranges from 20 to 80. The scores were also converted into percentages; scores below 50% were categorized as poor, while those above 50% were good attitudes. Section D consisted of 15 questions on the practice of pain management using Likert scales of 1-4 with a total score of 60 points.

After ethical permission, directors of Nursing Services in the two selected hospitals were contacted; three nurses were chosen and trained in each facility to serve as research assistants in the data collection process, which involved three stages (pre-intervention, intervention, and post-intervention). Nurses working in different units of the hospital participated in the study. For this study, the two groups took the same pretest to establish the initial baseline data, thus identifying gaps in knowledge, attitude, and practice level of pain management in both hospitals. After identifying the gaps, both groups undertook a short training session on pain management. Finally, the two groups took the same post-test four weeks after the training. This was conducted to determine the effectiveness of the intervention and to know if the intervention actually made a difference in the knowledge, attitude, and practices of the nurses in the management of patients' pain. The comparative effectiveness of the intervention was then evaluated by comparing the observed transformation between the pre-and post-tests of the two groups. Data were collected from October to December 2021.

After agreeing on a date with the directors of both hospitals, the researchers met with the targeted participants, and written informed consent was obtained from the nurses during the first visit. The researchers discussed the study’s objectives, goals, and benefits and four stages of training with the participants. During the first session, which was the first week, participants were introduced to the study and the pretest was administered. This was conducted to examine the participants' preintervention knowledge of pain assessment and non-pharmacological as well as pharmacological pain management.

The second session, which was the teaching/training session, spanned from the second to the fifth week. The teaching session involved extensive training on the knowledge, attitude, and management of pain. The researchers also talked about the different types of pain, such as incident, breakthrough, and procedural pain, and how to employ rescue doses to change daily opioid doses based on rescue dose requirements. Participants were also trained on how to recognize and treat different types of pain. Audio-visual lectures were provided in the hospital conference halls. The entire session was divided into four modules that were taught on separate occasions to ensure extensive teaching as well as communication of ideas between the researcher and participants. Each session lasted for an average of two hours. To facilitate the learning process, the training is also comprised of specific methods such as role-playing, simulation, and case studies.

The third session (the sixth and seventh week) was about the importance of pain evaluation (using various instruments) and the World Health Organization analgesic ladder, which was used to review and discuss pharmacological treatment options for mild, moderate, and severe pain. The non-pharmacological pain management techniques were also covered, the session also lasted for about 2 hours, and participants’ questions were adequately answered.

The final session occurred in the eighth week. Previous teaching and exercises were extensively reviewed. In this session, the researcher also educated the participants on various types of adjuvants/co-analgesics, analgesic side effects and toxicity, pain therapy in children, the elderly, patients with dementia, sickle cell disease, addiction, and dependence. Finally, the evaluation session occurred between the eleven and twelfth weeks, and postintervention tools were used to assess the training session's success in terms of the participant's knowledge, attitude, and pain management practice.

Data were analyzed using the Statistical Package for Social Science (SPSS) software version 25. Data were presented using descriptive and inferential statistics. Comparisons of means were carried out using the One-Way Analysis of variance, while tests of associations were carried out using Chi-Square.

Before commencing the study, ethical approval was obtained from the Research and Ethics Committee of the Kwara State Ministry of Health, Ilorin, Kwara State, Nigeria, with protocol number: ERC/MOH/2021/12/012. In addition, approval was obtained in each of the respective hospitals where the study was carried out. In addition, throughout the research process, steps were implemented to ensure the study complied with the ethical principles of research. Participants’ rights were explained to each participant, and none was coerced into participating in the study. Moreover, informed consent was obtained from all participants willing to participate in the study. All the information provided by the participants was treated with utmost confidentiality, and the participants’ identities were not revealed to anybody at any point in time throughout the period of study.

The socio-demographic profile of the study participants revealed that most participants were within the age range of 21-30 (56.6%) in Hospital A and 31-40 years (42.1%) in Hospital B. The majority of the participants were females (84.3% in Hospital A and 81.6% in Hospital B). With respect to the area of nursing care, 34.9% of them worked in medical or emergency/trauma wards in Hospital A, while 26.3% and 21.1% of participants in Hospital B worked in medical and emergency/trauma wards, respectively (Table 1).

In addition, over 80% of the participants in both hospitals had either a diploma or bachelor’s degree in nursing, while only a few had postgraduate qualifications. Moreover, 22.9% and 44.7% were principal nursing officers. In the case of work experience, 43.4% of participants in Hospital A had 1-5 years of working experience, while 29.5% of those in Hospital B had 11-15 years of experience (Table 1).

With respect to participants’ knowledge level of pain management, 45.8 and 21.1% of participants had adequate knowledge of pain management before the intervention, but after the intervention, 100 and 92.1% of participants had adequate knowledge of pain management in Hospitals A and B, respectively. Among the socio-demographic characteristics of the participants in Hospital A, only gender was observed to be significantly associated (X²= 6.022, p= 0.014) with knowledge level before intervention. There was, however, no significant association between the demographic profile of participants in Hospital B and the knowledge level of pain management before intervention. In both hospitals, none of the demographic characteristics of the participants showed any significant association with the knowledge level of pain management after intervention (Tables 2 and 3).

Generally, almost all participants in both hospitals (98.8% in Hospital A and 97.4% in Hospital B) were observed to have a good attitude towards pain management before and after the intervention. None of the participants’ demographic characteristics showed a significant association with attitude to pain management. This observation was irrespective of the participants in the two hospitals used for the study (Tables 4 and 5).

With respect to the practice of pain management, the study findings revealed that almost all participants had good practice. This observation was irrespective of pain management practice in the two hospitals and the practice before and after the intervention. Furthermore, none of the demographic characteristics of the participants in both hospitals was observed to be significantly associated with pain management practice (Tables 6 and 7).

Generally, the knowledge scores of the participants were observed to be significantly lower before the intervention compared to those after the intervention. This observation was irrespective of the two hospitals (p= 0.000). With respect to knowledge scores before the intervention, those of participants from ‘Hospital A’ were significantly higher than those from ‘Hospital B’. In the case of scores after the intervention, although the average score of participants from ‘Hospital B’ was higher than that of those from ‘Hospital A’, the difference was insignificant (Table 8).

As shown in Table 6, the scores of attitudes toward pain management by the participants did not differ significantly before and after the intervention. This observation was irrespective of the hospital. In addition, the attitude scores before and after intervention did not vary significantly between the two hospitals (Table 8).

A comparison of the pain practice scores of the participants revealed no significant difference between practice scores before and after the intervention. This observation was irrespective of the respondents in both hospitals. However, practice scores were significantly higher for respondents from ‘Hospital B’ than those from ‘Hospital A’ (Table 8).