Nurses’ pain management practices for admitted patients at the Comprehensive specialized hospitals and its associated factors, a multi-center study

Background Pain is the most common challenge that most hospitalized patients complain of and is influenced by several patients, nurses, and institutional-related factors. Most studies in Ethiopia on pain were focused on surgical illnesses only. Objective To assess nurses’ pain management practice and associated factors for admitted patients at Comprehensive Specialized Hospitals. Methods and materials : A multi-center institution-based cross-sectional study was conducted at the five randomly selected Comprehensive Specialized Hospitals of the Amhara region from May 01 to June 01, 2022. A multi-stage sampling method was employed to select a total of 430 nurses and patients for whom the nurses were responsible. Data were collected using standard self-administered, structured, and checklist questionnaires from nurses, patients, and patients’ charts respectively. The modified Bloom’s criteria categorized the overall practice as good, moderate, and poor. Data were checked, coded, and entered into Epi-Data version 4.6 and exported to SPSS version 25. An ordinal logistic regression model was applied, and variables with a p-value < 0.05 with a 95% CI in the multivariable analysis were considered significant. Results The study evaluated the pain management practices of 430 nurses and only a quarter had good pain management practices. Those nurses with first degrees and above education level (AOR = 2.282) and who attended in-service training (AOR = 2.465) were found to have significantly higher pain management practice. Expected though patients with painful procedures (AOR = 5.648) and who had severe pain (AOR = 2.573) were receiving better pain management practices from their nurse care provider. Nurses working in the institutions with a pain-free initiative focal person (AOR = 6.339) had higher pain management practices. Conclusion and recommendation : Overall, the majority of nurses had poor pain management practices. Higher educational levels, in-service training, and assigning a pain-free focal person had an impact on pain management services. Patients with higher pain levels and painful procedures were getting better attention. Hospital administrations need to provide due attention to the pain management of hospitalized patients by providing in-service training and educational opportunities to improve the capacity of nurses. Patients would be benefited considerably if hospitals focus on assigning focal persons for advocating regular pain management for admitted patients regardless of their pain level. Supplementary Information The online version contains supplementary material available at 10.1186/s12912-023-01528-x.

Nurses' pain management practices for admitted patients at the Comprehensive specialized hospitals and its associated factors, a multi-center study

Introduction
The International Association for the Study of Pain (IASP) defined pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage [1].It is a subjective sensation caused by high-intensity stimuli triggering pain receptors in the skin, muscle, bone, and other tissues [2].Pain is now recognized as the "fifth vital sign, " which is assessed and managed regularly.If it is not treated timely and properly, pain affects almost every aspect of the life of the patient as well as the family [3,4].
Pain management practices refer to those activities done by health care providers, particularly nurses to relieve those who are in pain, which includes pain assessment using organizationally accepted tools, providing pharmacological and non-pharmacological pain management, and evaluation of pain management by reassessing patients' status.It is effective when patient outcomes, satisfaction levels, and quality of life are all improved [5,6].
Despite regular pain assessments and management protocols, the burden of pain among hospitalized patients remains significant.More than half of admitted patients complain of pain during most of their hospital stay (7)(8)(9)(10).Certain studies in different areas including Ethiopia revealed that nurses' pain management practice for hospitalized patients is inadequate [11][12][13][14].
There have been different patient-related factors influencing the pain management practices of nurses according to the findings of different studies in Ethiopia.For instance, patients' gender was one of the factors in which being female was about five times higher level of pain as compared to males which in turn influenced the pain management practices of nurses according to Debre Tabor of Ethiopia (15).A study conducted in Gondar revealed that pain severity had a significant association with the nurses' pain management practices [16].
According to the findings of different studies; knowledge of nurses to the pain assessment and management, age, educational qualification, work experience, and inservice training were nurse-related factors that had significant association with the pain management practices of nurses for hospitalized patients [11,13,[17][18][19][20]. Nurseto-patient ratio was an institutional-related factor which was indicated by insufficient number of patients for the number of patients was associated with inadequate pain management practices [13,20,21].
The factors; particularly the patient-related factors that influence the nurses' pain management practices were not studied well in Ethiopia.Another limitation of Ethiopian studies on pain management practices is they frequently rely on surgical patients, particularly postoperative ones.Therefore, the current study aimed to investigate pain management practices and factors that act as barriers or facilitators to nurses' pain management practices which subsequently increase the burden of pain among hospital admitted patients.

Study design and setting
An institutional-based multi-center cross-sectional study was conducted among nurses working at five randomly selected referral hospitals (Tibebe-Ghion, Felege-Hiwot, Debre-Tabor, Debre-Markos, and Debre-Berhan) of Ethiopia and patients admitted at these hospitals from May 01 to June 01, 2022.

Inclusion criteria
All nurses working in the inpatient departments of the selected Hospitals during the study period were considered in this study.
All patients admitted to the inpatient departments of the selected Hospitals during the study period were considered in this study.

Exclusion criteria
Nurses on annual leave or sick leave during the data collection period were excluded from the study.Patients with critical illness and impaired level of consciousness were excluded from the study.

Sample size and sampling
The sample size was determined using a single population proportion formula by taking a proportion of 23.53% from Jimma University Medical Center (JUMC) [11].By adding a 5% non-response rate to the initial sample, the sample size becomes 291.The sample was again multiplied by a design effect of 1.5 yielding the final sample size of 437, which is the number of nurses who were included in the study.
The same number (437) of patients were included in the sample to assess pain intensity directly from the patients and the pain management practices of nurses from the corresponding patient's chart.
Nurses play a key role in the management of patient's pain and they are in charge of patient care during their scheduled working hours as well as the patients that they are assigned for.The pain management practice of nurses was evaluated from the records of the patient's charts for whom he/she was responsible.That is why an equal number of nurses and patients were intended to be studied.Before selecting a nurse for the study, the bed numbers that he/she was responsible for were determined.Then, one of the patients for whom the nurse is responsible was chosen at random so that the chart could be reviewed and patient-related characteristics could be evaluated.From the patients' charts, data about administered medications such as anti-pains and vital signs incorporating the latest pain score were collected.The charts were reviewed about four hours back to evaluate the corresponding nurses' pain management practices.During this time the nurse involved in this study was assumed to be responsible for the care and records that have been done to the patients participated.Because nurses are only assigned to a certain number of patients for a specified amount of time in hospital settings.

Sampling procedure
A multi-stage sampling technique was employed to select the study samples.First of all, a list of all comprehensive specialized hospitals had been obtained from the Amhara health bureau.Five hospitals (Tibebe-Ghion, Felege-Hiwot, Debre-Tabor, Debre-Markos, and Debre-Berhan) were selected randomly from the eight comprehensive specialized hospitals.Following that, each institution provided the number of nurses employed in both the outpatient and inpatient departments who rotated between the two settings.Then the calculated sample size was proportionally allocated to the size of the population in the selected hospitals.Five inpatient departments (Medical, Surgical, orthopedics, Oncology, and Maternity) were selected in each hospital, and nurses working in those departments again were allocated proportionally based on their number.Finally, simple random sampling by lottery method was employed to select the study samples.
The total number of patients admitted per month was obtained from each department's Health Management Information System (HMIS) database.A total of 437 patients with their hospital patient charts were selected from the selected units of each hospital to determine pain intensity.Before selecting a nurse for the study, the bed numbers he/she was responsible for were determined.Then, one of the patients that the nurse is responsible for was selected.The details of the sampling procedure is presented below Figure 1.

Study variables
Dependent variable Nurses' pain management practice.

Independent variables
Independent variables are categorized into three groups, which include:-Patient-related variables Socio-demographic variables (age, sex, educational status, marital status, religion, occupation) and Clinical factors (painful procedure, latest pain score record, type of analgesics, and pain intensity).
Nurse-related variables Socio-demographic characteristics (age, sex, educational status, work experience, and working unit), and knowledge were nurses-related variables.
Institutional-related variables In-service training, nurse-to-patient ratio, written policy for pain as the "5th vital sign", pain assessment tools, pain management guidelines, number of working shifts, and pain-free hospital initiative implementation focal person were institutionalrelated variables.

Good pain management practice
According to the checklist evaluation, nurses with scores ≥ 75% were regarded as having good pain management practice [12,24,25].

Moderate pain management practices
Based on the checklist evaluation, nurses with scores between 50 and 74.99% were considered as having moderate pain management practices [12,25].
Poor pain management practice Based on the checklist evaluation, nurses with scores less than 50% were considered as having poor pain management practices [12,24,25].
Adequate nurse to patient ratio When a nurse serves ≤ 6 patients in all units other than the Intensive Care Unit (ICU) which is based on Ethiopian Standard Agency (ESA) for Comprehensive hospitals [26,27].
Good knowledge Those nurses who scored above or equal to 80% of the knowledge questions were regarded as having adequate knowledge about pain management [24].
Painful procedure Any procedure that could reasonably be anticipated to result in more than mild or transient pain and/or distress in the patients to whom it is applied [28].

Data collection tools and procedures
Data were collected using a self-administered questionnaire from nurses and an interviewer-administered questionnaire from patients as well as a checklist for chart review and observation.Some parts of the questionnaire prepared for this study whereas, the other parts were taken from previous studies.An 8-item binary scoring (yes/no) data extraction checklist was used to evaluate nurses' pain management practices.Those who scored ≥ 75% [6] were regarded as a good pain management practice, those who had a score of 50-74.9%(4-5) were regarded as moderate practice, and those who had a score of < 50% [4] poor pain management practices.The tool was adapted from the Pain-Free Initiative Implementation Manual and different studies in Ethiopia [12,24].A 12-item with binary scoring (yes/no) questionnaire was used to assess the level of nurses' knowledge on pain management.Those nurses who scored ≥ 80% (10) of the knowledge questions were regarded as having good knowledge.It was adapted from different Ethiopian studies (12,24).The WHO pain ladder was used as a tool to assess the pain intensity of patients and was reported as no pain (zero), mild pain (one-three/1-3), moderate pain (four-six/4-6), and severe pain (seven-ten/7-10) (23,29).A 6-item checklist was used to determine institutionalrelated factors to nurses' pain management practices [11,13,30].The questionnaire also contains the socio-demographic characteristics of nurses and the sociodemographic and clinical characteristics of patients.Four who were not working at study hospitals collected the data after a full day training.

Data quality control
The data quality was maintained by using a carefully designed questionnaire collected by well-trained data collectors and supervisors.Three diploma nurses and two BSC nurses were recruited to collect and supervise the data collection process in each hospital.The data collectors and supervisors were trained for one day on procedures, techniques, and ways of collecting the data.Data were checked for completeness and consistency.A pretest was conducted at the University of Gondar Comprehensive Specialized Hospital (UoGCSH) on 22 nurses.

Data analysis and processing
Data were checked, coded, and entered into Epi-Data version 4.6 and exported to SPSS version 25 for analysis.Descriptive statistics such as frequency distribution, proportion, and percentages were done.The extent of multi-collinearity between independent variables was checked using variance inflation factor (VIF) and tolerance and the values of both statistics for all independent variables were within an acceptable range (VIF < 10 and Tolerance > 0.1).Regression model assumptions were checked using a test like the Hosmer & Lemeshow goodness of fit with a P-value (Pearson chi-square = 0.096 and Deviance-0.776).The proportional odds assumption was checked with a test of parallel lines with a P-value ≥ 0.05 which is an indicator for the fulfillment of the assumption.A P-value for the test of parallel line for this study was 0.421.
A bivariable ordinal logistic regression model was fitted for each explanatory variable.Accordingly, those variables having a p-value ≤ 0.25 in the bivariable regression analysis were taken as the candidates for the multivariable ordinal logistic regression model.In multivariable regression, variables with a P-value of less than 0.05 with a 95% CI were considered statistically significant.

Socio-demographic characteristics of nurses and patients
The study included 430 nurses in total, with a response rate of 98.4%.Over half 228 (53%) were males and the mean age was 30.44 with a standard deviation (SD) of 4.311.Majority 374 (87%) of nurses had an educational qualification of degree and above.Majority 206 (47.9%) of nurses had a work experience of 5-10 years (mean = 7.59, SD = 4.05).Only 58 (13.5%) of nurses had attended in-service training regarding pain management in the past two years (Table 1).
In this study, 430 patients were interviewed, their pain level was assessed using NRS, and the charts of the corresponding patients were reviewed.The majority were females 260 (60.5%) and the mean age was 45.32 years with a range of 18-86 years.Regarding the educational status of the patients, 148 (34.4) were unable to read and write, while 132 (30.7%), 67 (15.6%), and 83 (19.3%) had primary, secondary, and diploma and above educational levels respectively.Most of the patients were farmers (25.3%) and housewives (35.1%) (Table 1).

Knowledge of nurses toward pain management practices
The mean nurses' knowledge of pain management score was 8.84 with a standard deviation of 1.92.The most correctly answered item was "Pain should be assessed before and after administering anti-pain drugs" with a 98.8% correct response rate.The question "Acetaminophen and non-steroidal anti-inflammatory agents are effective analgesics for severe pain" was the least answered item with only a 24% correct response rate.Of all nursing professionals, the majority 273 (63.5%) had poor knowledge of pain management practices ((Table 2).

Clinical characteristics of patients
Regarding the clinical characteristics of patients, only 207 (48.1%) of all patients' pain was recorded within 24 h before the chart review.More than half of 238 (55.3%) patients had undergone painful procedures.When assessed using NRS, 351 (81.63%) had reported mild to severe pain with a median and IQR of 3 and 4 respectively ((Table 3).

Institutional-related factors for pain management practices
Five of the Comprehensive Specialized Hospitals of the Amhara region were included in the study.4 (80%) of the five hospitals had two working shifts and 3(60%) hospitals had a 1 to 6 nurse-to-patient ratio for working hours.Of those hospitals, 4 (80%) had a written policy for pain as the 5th vital sign, 3 (60%) had pain management guidelines, and only one (20%) of the institutions had a pain-free hospital initiative focal person.

Pain management practices of nurses
About 222 (51.6%) of the charts revealed pain was assessed and recorded at least once within the past four hours by the nurses.Among those nurses evaluated by chart review of their particular client, only 152 (35.3%) had performed a regular pain assessment during their stay at the workplace.Regarding the level of practice, only 111 (25.8%) of nurses had good pain management practices while 118 (27.4%) and 201 (46.7%) had moderate and poor pain management practices respectively ((Table 4).

Factors associated with Nurses' pain management practice
A total of 23 variables were included in the model of bivariate ordinal logistic regression level.On the multivariable ordinal logistic regression, five (5) variables become significantly associated with pain management practices at a P-value less than 0.05.These variables are nurses' educational level, in-service training, patients with the painful procedure, pain intensity, and assigned pain-free initiative implementation focal person in the institutions.
The educational status of nurses was found to have a statistically significant relationship with the pain management practices of nurses.The odds of being in a higher level of pain management practice was 2.282 times (AOR-2.282,with 95%CI: [1.181-4.41])higher on average for those having an educational qualification of degree and above as compared to those having a diploma.
In-service pain management training was found to be significantly associated with pain management practices.The odds of being in a higher category of pain management practices was 2.47 times (AOR = 2.465 with 95%CI:[1.317-4.614])higher on average for those nurses who attended an in-service pain management training as compared to those who didn't attend.In other words, those nurses who attended pain management training had a 71% probability of having a higher level of pain management practices.
From the patient-related factors, patients with painful procedures become a significant predictor of nurses' pain management practices.The odds of being in a higher category of pain management practice was 5.648 times (5.648 with 95%CI: [3.237-9.856])higher for those nurses who provide care for patients with painful procedures as compared to the contrary group.Those nurses who follow those patients with painful procedures had an 85% probability of falling in the higher category of pain management practices.
The other patient-related significant variable was pain intensity.The odds of being in a higher level of pain management practice was 2.573 times (2.573 with 95%CI: [1.35-5-4.899])higher on average for those nurses who follow patients with severe pain as compared to those who follow patients with mild pain.Similarly, the odds of being in a higher category of pain management practice was 3.236 times (3.236 with 95%CI: [1.771-5.914])higher on average for those nurses who follow patients without pain as compared to those who follow patients with mild pain.
From institutional factors, a pain-free initiative implementation focal person was found to have a statistically significant relationship with the nurses' pain management practices.The odds of being in a higher category of pain management practice was 6.34 times (6.339 with 95%CI: [3.611-11.13])higher on average for those nurses working in the institutions having a pain-free hospital initiative implementation focal person as compared to the contrary group.Those nurses working in the hospitals having a pain-free hospital initiative implementation focal person had an 86.4% probability of having a higher level of pain management practices (Table 5).

Discussion
This study aimed determine nurses' pain management practices and associated factors for admitted patients at the Amhara region comprehensive specialized hospitals.The study indicated that only 111 (25.8%) with a 95%CI [0.217, 0.302]) of the nurses had good pain management practices, while 118 (27.4%) and 201 (46.7%) had moderate and poor pain management practices respectively.The finding of this study was consistent with the study conducted at Hawassa Referral Hospital (24.4%) [12].However; it was lower compared to the studies done at the Federal Hospitals of Addis Ababa (56.5%), governmental hospitals of Harari region and Dire Dawa city administration (33.6%), and Jimma University Medical Center (JUMC) (76.47%) [11,13,31].
This gap could be explained by the difference in sample size, as the sample sizes in the earlier research were generally less than those in this study.On the other hand, the discrepancy may have arisen from the use of an objective   tool to evaluate nurses' pain management practices in the current study.The disparity in the study populations may be another factor contributing to the discrepancy.The studies done in Addis Ababa (13) was on the practices of nurses for critically ill patients and that of Jimma (11) and Dire Dawa (31) were for post-operative patients.
Whereas the current study was in all adult inpatient units as well as multicenter.Patients with critical illnesses typically need to be closely monitored on a regular basis, which forces nurses to pay attention to their patients' pain levels and treat them appropriately.On the other hand, compared to patients without critical illness, postoperative patients, critically sick patients, and those with more severe pain tend to report their pain status and could be managed.However, pain management might not be as effective as it could be in the case of patients who are unable to report and have critical illnesses [16,32,33].This study was multicenter which was in contrast to the studies of Jimma, Hawassa, and Dire Dawa which might result in a difference in the level of nurses' pain management practices.The other reason that might have brought the difference in the level of nurses' pain management practices could be the difference in the provision of pain management training.For example, only 13.5% of professionals attended pain management training in this study which was lower compared to the finding of Jimma in which 23.5% of professionals had attended in-service pain management training which might have resulted in the difference in the level of pain management practices of nurses.
The result of this study was relatively higher as compared with the study done in Rwanda in which only 2% of the study participants had good pain management practices (14).The possible reason for the discrepancy could probably be attributed to the educational level of nurse professionals in Rwanda in which only 4% of nurse professionals had an educational qualification of degree which is far less than the current study in which 87% had a degree and above level of education.When patients' pain is not managed sufficiently, they may experience serious negative effects, including socioeconomic, clinical, and emotional ones.It could lead to chronic persistent pain, unexpected readmission, and delayed recovery.Regular assessment and management of pain are essential for a better patient outcome [34].
As expected the educational status of nurses was found to have a statistically significant relationship with the nurses' pain management practices.The odds of being in a higher level of pain management practice was nearly 2.3 times higher on average for those with educational qualifications of a degree and above than those with a diploma.The possible reason would be that nurses with a degree and higher levels of education typically have better knowledge and comprehension of pain, its assessment, and management mechanisms, which in turn leads to better pain management practice.A higher level of education has an impact on nurses' pain management practices, i.e. if someone is not equipped with the necessary knowledge and skills important to manage pain, patients will not get an appropriate pain management service from him/her and end up suffering from pain.The knowledge and clinical skills of professionals will be greatly enhanced by advanced nursing education, which will lead to better pain management for patients.Patients would also benefit when they get attention to their pain from skillful professionals throughout their hospital stay.The finding was supported by the studies conducted in Hawassa and Rwanda [12,14].
This study indicated that in-service pain management training was found to be significantly associated with pain management practices.A higher category of pain management practice was more likely for nurses who participated in in-service pain management training compared to those who did not in the previous two years, by a factor of 2.47.This is because in-service training can provide nurses with more knowledge, insight, and understanding of patients' pain, which may enhance their pain management practices.In other words, it serves to update the nurses' pain management knowledge and practical skills and improve the best practices for fulfilling their responsibilities in the regular assessment and management of pain for patients.It also plays an essential role in improving the standard of inpatient pain management practices [35,36].This finding was supported by the study conducted at Jimma University, the Federal Hospitals of Addis Ababa, and the two public hospitals in the Oromia region [11,13,37].
In this study, patients with painful procedures become a significant predictor of nurses' pain management practices.Nurses who care for patients with painful procedures had 5.65 times higher odds of falling into a higher category of pain management practice than the opposite group.This could probably be because painful procedures increase the likelihood of high-intensity of pain that drives patients to report their pain to their care providers.Nurses who take part in caring for patients with painful procedures might be able to reduce the suffering of patients with procedural pain which put them in the higher category of pain management practices [38].
The other patient-related significant variable was pain intensity.Nurses who care for patients with severe pain had, nearly 2.6 times higher odds of practicing higher levels of pain management than nurses who care for patients with mild pain.The possible reason might be that patients with severe pain are more likely to report their pain status and get more attention from their healthcare providers.According to findings of studies in Gondar [16] and Central Africa [33], patients with severe pain are managed more frequently, relatively early, and regularly than those with mild and moderate ones which puts nurses who follow patients with severe pain in the higher category of pain management practices.
From institutional factors, a pain-free initiative implementation focal person was found to have a statistically significant impact with the nurses' pain management practices.The odds of being in a higher category of pain management practice was nearly 6.34 times higher for those nurses working in the institutions having a painfree hospital initiative implementation focal person as compared to the contrary group.The possible reason behind this could probably be that nurses may improve their pain management activities if they are guided by trained professionals who feel responsible for the proposed area of practice.Additionally, the pain-free hospital initiative would be implemented when there is a responsible individual who can advocate a regular patient assessment and management of patients with any pain level.Those who are assigned as focal persons have vital importance in the improvement and continual provision of pain management for hospitalized patients.This could be achieved by preparing a work plan, evaluating pain management activities, and coordinating patient education programs [36,39].

Conclusions and recommendation
This study found that the majority of nurses had poor pain management practices for admitted patients.Nurses with higher education level and having in-service training were providing better pain management to their admitted patients.Patients with painful procedures and higher pain intensity were receiving better pain management from their nurse care provider.From several institutional factors, the assigned pain-free hospital initiatives implementation focal person was another factor that remained significant to the pain management practices of nurses.
Hospitals administrations need to provide due attention to the pain management of hospitalized patients by providing in-service continuous training to improve the capacity of the health care providers.It would improve the service of pain management for admitted patients if there is a focal person for a pain-free hospital initiative implementation who can advocate a regular pain assessment and management for each admitted patient regardless of their pain severity.Finally, future researchers would get better findings if they perform an observational follow-up study to determine the pain management practices of nurses.

Limitations of the study
The study used only patients' charts to evaluate nurses' pain management practices for their corresponding patients as it was difficult to observe and follow all of the nurses within the specified period.Social desirability bias during data collection was another drawback of this study.The other limitation was that it didn't include the pain management practices for patients who were with critical illness and impaired level of consciousness since the pain assessment tool (NRS) was not suitable for them.

Fig. 1
Fig. 1 Schematic presentation of sampling procedure on pain management practices and associated factors of nurses for admitted patients at the Comprehensive Specialized Hospitals of the Amhara region of Ethiopia, 2022

Table 1
Socio-demographic characteristics of nurses (n = 430) and patients (n = 430) at Amhara Region Comprehensive Specialized Hospitals 2022

Table 2
Knowledge of nurses toward pain management practices for admitted patients (n = 430) at Amhara Region Comprehensive Specialized Hospitals, 2022 G.C.

Table 4
Pain management practices nurses for admitted patients (n = 430) at Amhara Region Comprehensive Specialized Hospitals, 2022

Table 5
Bivariable and Multi-variable ordinal logistic regression analysis to factors associated with pain management practices of nurses for admitted patients (n = 430) at Amhara Region Comprehensive Specialized Hospitals, 2022