This study employed a quantitative descriptive cross-sectional design to examine nurses’ knowledge of hypertension care and associated factors. The study was conducted in the Tamale Metropolis, the capital of the Northern Region of Ghana. Specifically, it was done at the three major government hospitals, namely Tamale Teaching Hospital (TTH), Tamale Central Hospital (now the Northern Regional Hospital) (TCH) and Tamale West Hospital (TWH). Tamale Teaching Hospital is a tertiary-level healthcare facility, and the other two provide secondary care. These hospitals were selected because they serve as the main referral centres in the Northern Region. Also, because hypertension and other chronic diseases management in Ghana are mainly anchored in secondary level and above hospitals.

The study population comprised nurses working in the outpatient department (OPD), hypertension clinic and the medical wards of the selected hospitals. Newly qualified rotation and newly employed nurses who have worked for less than 6 months were excluded. The estimated average population from which the sample was drawn was approximately 410 nurses (TTH: 230 [56%], TCH: 90 [22%] and TWH: 90 [22%]). The sample size was initially calculated using a precision-based method, based on a population size of 410 nurses, a 95% confidence interval and a 5% margin of error, yielding in an estimated sample of 199 respondents. Respondents were proportionally allocated across hospitals. However, rather than limiting recruitment to this number, the census approach was adopted. To maximise response rate and enhance representativeness, all eligible nurses across the participating hospitals were invited to participate.

The study instrument was adapted from two validated instruments (Campbell et al. 2017; Serrat-Costa et al. 2022). The original instrument of Campbell et al. (2017) has 16 knowledge questions (with question 4 having six sub-questions and question 10 having two sub-questions, making a total of 22 questions), while Serrat-Costa et al. (2016) has 23 questions. These two instruments were integrated, and some questions were removed for contextual reasons, resulting in 34 knowledge questions. The 34-item questions assessed knowledge under three domains: ABPM (Section A), hypertension diagnosis (Section B) and hypertension management (medication and lifestyle) (Section C). The questions included multiple choice questions (including choosing all that apply; yes, no and I don’t know) and short answers.

Before using the survey instrument, Campbell et al. (2017) indicated that the survey’s specific questions need to be examined for their local and national applicability. In the instrument of Serrat-Costa et al. (2022), a Cronbach’s α greater than 0.7 was obtained. In this study, a reliability test was done for the individual domains, resulting in a Cronbach’s alpha of 0.51 for the BP measurement domain. The hypertension diagnosis and hypertension management domains resulted in Cronbach’s alphas of 0.49 and 0.51, respectively. However, when all three domains were computed, the overall reliability analysis for the instrument yielded a Cronbach’s alpha of 0.66, indicating a moderate internal consistency. Although it is slightly below the recommended threshold, this value could be attributed to the high heterogeneity of the content (Izah, Sylva & Hait 2023; Tavakol & Dennick 2011).

Campbell et al. (2017) involved hypertension and primary care experts to determine the face validity of the knowledge, attitude and practice (KAP) instrument; it was also pilot-tested among primary healthcare professionals (sample size not stated). The same instrument was used in two studies (Myanganbayar et al. 2018, 2019). The second instrument (Serrat-Costa et al. 2022) also met all Rasch-Item Response Theory (Rasch-IRT) model criteria: item- and person-fit measurement, unidimensionality, local independence, invariance, targeting and reliability. Face validity is defined as the degree to which test respondents view the content of a test and its items as relevant to the context in which the test is being administered (Allen, Robson & Iliescu 2023), while content validity refers to the relevance, comprehensiveness and comprehensibility of the content of an instrument (Mokkink et al. 2025). In this present study, content validity was examined by involving subject experts (a cardiologist and a senior nurse in hypertension care) to assess the appropriateness and relevance of the test and items as they appear to the respondent. They recommended that some questions were outside the scope of nursing practice and should be removed. The two co-authors (also nurses) thoroughly reviewed the questionnaires to determine their relevance to the test purpose. A pilot test was conducted to assess item ambiguity, ease of answering and applicability to this study as a means of determining the face validity (Allen et al. 2023). Because the adapted instruments are existing validated instruments, they were examined to determine their applicability in the Ghanaian context. The pilot test resulted in the removal of five questions from the instrument because most respondents complained about the ambiguity of those questions.

The Principal Investigator (PI, SPA) engaged with the nurse managers of the respective study sites to introduce the study to the eligible nurses. The PI (SPA) is a staff member of the TTH and was therefore able to approach the eligible nurses. Although the principal PI is a staff member of TTH, she does not hold any managerial role over the eligible nurses. Formal permission to engage nurses was obtained from the nurse managers of the various units, who acted as the institutional gatekeepers. Eligible nurses were informed about the study either through unit managers or the PI and research assistant, and participation was entirely voluntary. The research team approached the identified nurses at the close of their work shift to prevent interfering with their working hours. They were informed about the purpose of the study, and participant information sheets were provided for them to review for details and clarity. Upon agreeing to participate, they were given consent forms to sign. Paper-based questionnaires were self-administered and returned anonymously by dropping them into a box in the nurse managers’ office, which was retrieved daily by the PI to minimise any perceived coercion and to ensure confidentiality. Other respondents who were interested in learning more about the study chose to hand in the completed questionnaires to the PI and ask questions. Email addresses and personal identifiers were not collected through the online version of the questionnaires. Most respondents answered paper-based questionnaires, and some responded online via REDCap (Harris et al. 2009), which was shared with respondents via WhatsApp. Paper-based responses were later entered into REDCap by the PI and a research assistant; at every 10th entry of the questionnaire into REDCap, the PI and research assistant (RA) cross-checked to ensure accurate data entry. Ten responses were excluded for not meeting the inclusion criteria: (1) students and (2) newly qualified with less than 6 months of working experience in hypertension care; they were thus not entered into REDCap.

Data from REDCap were exported to the Statistical Package for the Social Sciences (SPSS) version 30 for analysis (IBM Corporation 2024). During data cleaning, four more responses were eliminated because of incomplete data. Results are presented using descriptive and inferential statistics. The continuous variables (age, years of practice experience and number of years in the present unit) were summarised as mean ± s.d. Categorical variables (gender, hospital name, ward type, nursing category, level of education, hypertension training and knowledge questions) were calculated as frequencies and percentages. Knowledge scores were first divided into three sections and calculated as the number of correct or desirable answers divided by the number of items in each section (8 items in section A and 13 items in sections B and C) surveyed and expressed as a percentage. The total knowledge score was the total number of correct or desirable responses divided by the total number of items surveyed (34 items) and multiplied by 100%. Knowledge level was further categorised as good, average and poor. A score of ≥ 75% meant good knowledge, a score of 50% to ˂ 75% represented average knowledge and poor knowledge was a score < 50%. This categorisation was done to help distinguish between high and moderate knowledge levels.

In the study of Myanganbayar et al. (2019), a score of ˂ 50% represented low knowledge and ≥ 50% represented good knowledge. The authors of the instrument indicated that there is no strict scoring when using the instrument; authors adopting the instrument should contextualise it (Campbell et al. 2017). The Chi-square test was used to test for associations between demographic variables and knowledge categories. A significance level of 0.05 was applied.

This study, part of a broader project, received ethics approval from the Biomedical Health Research Ethics Committee (BMREC) of the University of the Western Cape, dated 05/12/2024 (BM24/10/6), and the Ghana Health Service Ethics Review Committee (GHS-ERC), dated 26/05/2025 (GHS-ERC:01103/35). Upon receiving ethics approval, written permission was obtained from the study sites. Written informed consent was obtained from all respondents involved in the study. Participants were assured of anonymity and confidentiality; they were not required to share any personal identifiers. They were also assured that their participation was voluntary and they could decline from the study without explaining to the researchers their reason for not continuing.

A total of 266 (64.9% response rate) nurses working in the outpatient and medical inpatient units from three different hospitals participated in the study. The mean age was 33.16 ± 4.62 years; the ages were categorised into two: that is, 23–33 years and 34+, with 56.4% below 34 years. The mean years of experience was 7.04 ± 4.36 years, and years of practice in their present units were 3.23 ± 1.96 years.

Table 1 presents the categorical demographic data of respondents; the majority of the nurses were from the TTH, representing 61.7% (n = 164). Nurses were asked to indicate the type of ward they work in their respective hospitals; the responses included OPD, Male or Female Medical ward, Hypertension Specialist Clinic, TTH Polyclinic OPD and ward. These were categorised into inpatient and outpatient units; 200 (75.2%) work in inpatient wards, and the rest in outpatient units. Fifty-three per cent (n = 141) of respondents were females. The majority of the respondents are Registered General Nurses (Professional nurses) (81.2%, n = 216), and the rest are enrolled (16.2%, n = 43) and Nurse Specialists (2.6%, n = 7). Just over half of the respondents reported a Bachelor of Nursing qualification (50.4%, n = 134), 28.9% (n = 77) a Diploma in Nursing, 14.4% (n = 41) a certificate and a few (5.3%, n = 14) a postgraduate degree. Respondents were asked if they received any training on hypertension prevention and management, and over half (56.8%, n = 151) indicated they received no hypertension training.

Respondents’ mean knowledge score for ABPM was 82.42% ± 16.65%, indicating that they were well informed in this category (Table 2). Respondents knew that the patient must sit upright and quiet, with the arm supported at heart level (94.7%, n = 252) and that patients must sit quietly for 3 min – 5 min before BP is taken (94.4%, n = 251). Further, 87.2% (n = 232) knew that the wrong cuff size gives a false reading. However, only 47% (n = 127) knew that the use of ambulatory BP measurement is a cost-effective strategy for diagnosing hypertension (Figure 1).

Respondents’ knowledge of the diagnosis of hypertension was poor, with a mean of 35.05 ± 15.74 (Table 2). Figure 2 presents details of respondents’ understanding of hypertension diagnosis. Under the individual variables, 79.7% (n = 212) were well informed about what resistant hypertension means. About 45% (n = 120) of respondents knew the lowest level of SBP considered hypertensive in adults, and 50.4% (n = 134) knew the lowest level of DBP considered hypertensive. Their understanding of BP values in controlled BP was poor, 3.8% (n = 10) and 7.1% (n = 19) for SBP and DBP, respectively.

Their knowledge of masked hypertension and white coat hypertension (which is potentially caused by inaccurate home and clinic BP measurements, respectively) was also low, scoring 15% and 4.1%, respectively.

The third category was medication and lifestyle management (Figure 3). The mean knowledge score was 48.84% ± 16.26% (Table 2). Most respondents knew that lifestyle change is essential and effective in managing hypertension and preventing cardiovascular events (95.0%, n = 253). Respondents were also well informed that two or more antihypertensives are required by most hypertensive patients (82.0%, n = 218), and 78.2% (n = 208) knew that side effects are one of the barriers to medication adherence. Regarding lifestyle, however, nurses’ knowledge was low; only 42.9% (n = 114) knew that the recommended daily level of salt consumption for people who have hypertension is < 6 g. A few respondents were well informed that 1 kg weight loss in an overweight hypertension patient leads to approximately 1 mmHg reduction in SBP. Regarding the alcohol intake limit for hypertension patients, 29.7% (n = 79) and 32.3% (n = 86) were well informed that male patients should limit to ≤ 2 and ≤ 1 for female patients, respectively. With the recommended weekly physical activity and fruits and vegetables intake, 9.8% (n = 26) and 12.0% (n = 32), respectively, were well informed.

The overall knowledge score of respondents was 51.47% ± 11.76%. Nurses’ overall knowledge level was categorised into three: good knowledge, for respondents who had ≥ 75%, average knowledge, for those who had 50% to < 75% and poor knowledge, for scores less than 50%. Figure 4 presents nurses’ total knowledge level; 4.1% (n = 11) had good knowledge, 57.1% (n = 152) had average knowledge and 38.7% (n = 103) had poor knowledge of hypertension care.

Respondents’ overall knowledge of hypertension care was further categorised under domains for ABPM (Figure 1), diagnosis (Figure 2) and medication and lifestyle management (Figure 3).

Table 3 shows the association between nurses’ knowledge of hypertension care and selected demographic variables. Regarding hospital of practice, more than half of the nurses with good knowledge were from TWH (54.5%, n = 6). In the average knowledge group, the largest proportion was from TTH (65.8%, n = 100) and TCH (19.1%, n = 29). Among those with poor knowledge, the majority were from TTH (58.3%, n = 60). A statistically significant association was found between hospital and knowledge level (χ² = 11.508, degrees of freedom [df] = 4, p = 0.021)

Regarding the unit of practice, among nurses with average knowledge, most were in the inpatient unit (72.4%, n = 110) and a smaller proportion in the outpatient unit (27.6%, n = 42). Poor knowledge was more common among nurses in the inpatient unit (80.6%, n = 83) than in the outpatient unit (19.4%, n = 20). Good knowledge was observed in both units, with 63.6% (n = 7) in the inpatient unit and 36.4% (n = 4) in the outpatient unit. There was no statistically significant association between unit and knowledge level (χ² = 3.041, df = 2, p = 0.219). Good knowledge was divided as 45.5% and 54.5% between males and females, with no statistical differences (χ² = 11.16, df = 4, p = 0.907). Age categories demonstrated a significant difference (χ² = 9.926, df = 2, p = 0.007), with poor knowledge observed more in nurses aged 23–33 years, 68.0% (n = 70). The nursing category showed a significant association with knowledge (χ² = 17.266, df = 4, p = 0.002), with Registered General Nurses making up the largest share of the average knowledge group (88.8%) and the good knowledge group (63.6%). Although Nurse Specialists were few, they were distributed across all categories. Among nurses with good knowledge, 9.1% (n = 1) was a nurse specialist, and among average knowledge and poor knowledge groups, 2.6% (n = 4) and 1.9 (n = 2), respectively, were Nurse Specialists. Educational qualification was significantly associated with knowledge (p < 0.001). Good knowledge (54.5%) and average knowledge (66.4%) was observed more in nurses with a bachelor’s degree or higher, while 60.2% of nurses with qualifications below a bachelor’s had poor knowledge. Although years of practice were not significantly associated with knowledge (p = 0.104), the data showed that nurses with 1–5 years of experience were more likely to have poor knowledge (53.4%), while those with 6–10 years demonstrated good knowledge (45.5%). The number of years in the present ward was significantly associated with knowledge (p = 0.021). Among nurses with average knowledge, most had 1–5 years in their present ward (82.9%, n = 126), and among those with poor knowledge, the majority also had 1–5 years in the ward (91.1%, n = 94). Conversely, nurses with longer ward experience (6+ years) accounted for a higher proportion of those with good knowledge (36.4%, n = 4).

Finally, in-service hypertension care training was significantly associated with knowledge (p = 0.027). Among those with good knowledge, nearly half had received training (45.5%, n = 5). In the average knowledge group, half had received training (50.0%, n = 76). By contrast, in the poor knowledge group, most had not received training (67.0%, n = 69).

Our study comprehensively assessed nurses’ knowledge of hypertension care in the Northern Region of Ghana using validated tools to measure their current knowledge of ABPM, diagnosis and management. The findings will inform the content of the guidelines to be developed in the broader project.

Regardless of this study’s strengths, it is not without limitations. Because this study was conducted solely in the Northern Region, the findings may not be fully generalisable to the entire country. Also, the study conducted among urban nurses could be a limitation, raising concerns of selection bias. The relatively low Cronbach’s alpha for the knowledge test suggests minimal internal consistency, which may limit the instrument’s reliability. Therefore, the findings should be interpreted cautiously.

Further studies should go beyond knowledge to investigate how nurses’ knowledge is translated into actual practice, including ABPM and patient education on lifestyle modification and medication adherence. Another research direction is to conduct an interventional study by providing training on hypertension to assess knowledge acquisition and retention at different time points.

The findings of this study, therefore, call for the design of a tailored continuous professional development programme on the prevention, diagnosis and management of hypertension for nurses. Hospitals and health institutions should institutionalise regular in-service training to ensure nurses’ knowledge is up to date with current hypertension guidelines. As part of the broader national non-communicable diseases (NCD) prevention and management strategies, the Ministry of Health, Ghana, should prioritise building nurses’ capacity to understand evolving hypertension guidelines and evidence-based practices.