Introduction

Clinical governance is a stewardship framework that was popularised by the UK’s National Health Service (NHS) and has been adopted by health institutions globally as an accountability framework for continuously improving the quality of care [1–3]. With its seven pillars; clinical effectiveness, risk management, patient experience and involvement, communication, resource effectiveness, strategic effectiveness, and learning effectiveness, this framework designates roles and responsibilities, and indicators on measures to be taken to ensure optimal patient outcomes [1].

In 2007, the South African National Department of Health (NDoH) submitted an amended policy of quality healthcare addressing challenges faced by hospitals during the provision of quality services [4]. This includes under and overuse of services, avoidable errors, lack of resources, inadequate diagnosis and treatment, inefficient use of resources, an inadequate referral system, disregard for human dignity, drug shortages, records not well kept and poor information, and poor health delivery systems [4]. The policy is in line with the South African National Development Plan (NDP) 2030, to promote change in the quality of healthcare goals and vision as supported worldwide and to reduce disease burden [5]. However, all of the intended policy goals have not been adequately addressed particularly in South Africa’s public health sector [6, 7] and established quality control programs have not been monitored. Overwhelming evidence is reported on a variety of issues that have a negative impact on healthcare quality; such as errors, delays in care delivery, efficacy and costs [7, 8].

To mitigate these setbacks, the NDoH as part of quality assurance, developed the National Core Standards (NCSs) against which service delivery by health establishments can be assessed [9]. The main focus of quality assurance (QA) is to oversee processes, which include adherence to standards and guidelines, or the arrangements and activities meant to safeguard, maintain, and promote quality of care [10]. The standards are not new or additional but combine existing policies and guidelines in a summary form illustrating mandatory organisational requirements and expectations for safe and decent care [9]. The framework is built on values and principles of compassionate care; and provides quality perspectives around health systems, healthcare and health outcomes [11, 12]. The department developed another framework in 2018, the Ideal Hospital Realisation and Maintenance Framework (IHRM-F). This framework is reported to be a critical strategy and intervention to facilitate improved health service delivery and strengthen health system effectiveness by capacitating hospitals to identify and address challenges [13]. The IHRM-F serves as a benchmark mechanism to improve the quality of care by setting and monitoring national norms and standards, improving systems for user feedback, increasing safety in health care, and by improving clinical governance [13]. However, with all these efforts in place, over the past decade, the South African health sector has noted an increase in medico-legal claims in both public and private hospitals [14–17]. Due to patients’ awareness of their constitutional rights, they are now more likely to make medical negligence claims. However, it is not known whether these increasing medico-legal claims in public hospitals can be linked to ineffective protocols, poor implementation of protocols or if frontline health workers are aware of the measures that are in place to improve the quality of care delivered in their hospitals.

Agreement is emerging that a bottom-up approach may be better—where the fundamental motivations of clinical staff are harnessed to drive quality improvement (QI) [18]. This is an appreciation that even though QI and clinical governance are often monitored at the highest level of governance of health facilities, frontline health workers have a significant role to play in ensuring that patient care is of the highest quality [19]. They can play an active role in conducting regular clinical audits, morbidity and mortality, and monitoring of trends [20]. Clinical governance outlines duties and responsibilities as well as indicators of the actions that should be followed to guarantee the best possible patient outcomes. Therefore, the purpose of this study was to ascertain how well-informed clinical staff were about the clinical governance protocols/tools that were utilised in four selected public sector referral hospitals in South Africa.

Methods

Data analysis

Data were captured and coded in Microsoft Excel and analysed with Stata version 18.0. Numerical data were explored for normality using the Shapiro Wilk test. The median and interquartile range (75^th percentile– 25^th percentile) were used to summarise participants’ age in years and years of experience. Categorical variables were summarised using frequencies, percentages and graphs. The Kruskal-Wallis test was used to compare the equality of median ages and years of experience of participants by hospital. Categories were compared using the Chi-squared test, but when the expected frequencies were <5 the Fisher’s exact test was used (Tables 1 and 2). Individual health workers’ responses were compared to the evidence supplied by that hospital on the presence or absence of the policies or practices and scored. Knowledge was generated based on the score being either below 50% or greater than or equal to 50% for the 12 questions assessed. Since scores were not normally distributed, the median and interquartile range (IQR) are used to summarise the knowledge scores between the facilities and compared using the Kruskal Wallis test. Binomial logistic regression models were used to determine associations of a positive response. The prevalence ratio (PR) is the measure of association used. To show the precision of estimates, 95% confidence intervals (95%CI) are reported, and the p-value used for statistical significance is, p-value≤0.05.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participants’ basic characteristics.

https://doi.org/10.1371/journal.pone.0312340.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Participants’ responses on protocols and practices in their hospitals.

https://doi.org/10.1371/journal.pone.0312340.t002

Results

Three hundred and seventy-seven (377) participants gave consent to participate in the study. Of these, 32.9% (n = 124) worked at SEH and 16.2% (n = 61) worked at Themba hospital. More than three quarters of the participants were female. Overall, professional nurses comprised 48.8% of the participants, and the median age was 40.2 years, with the oldest participant being 64 years old. Except for the years of experience, all other comparisons were statistically different between the four health facilities (p-value<0.008). Demographic characteristics are summarised in Table 1.

Confirmation of presence of protocols/tools and knowledge scores

Participants were asked about the existence of selected protocols, guidelines or practices within their hospitals (Table 2). Whilst 75.9% (277/365) of all respondents were affirmative that their hospitals had complaints management protocols, only 63.0% (63/100) of NMAH’s respondents and 88.6% (109/121) of SEH’s respondents were respectively positive about the presence of these protocols in their hospital. There was a statistically significant difference in reporting the presence of complaints management protocols across the different hospitals (p-value<0.0001). Likewise, 75.4% of all respondents were positive about the presence of adverse events protocols in their hospitals, 61.5% (64/104) and 84.7% (105/124) of NMAH’s and SEH’s respondents were respectively positive about the presence of this protocol (p-value = 0.007).

All individual staff responses were compared with the actual hospital’s reports; obtained from the CEO and Quality Assurance offices, on the presence or absence of the 12 protocols assessed (Table 3). Overall, 8.5% (32/377) of the participants got none or only one correct out of the twelve protocols; and 65.5% (247/377) got between two and five out of twelve correct. The differences between the hospitals were statistically significant (p-value = 0.002) with 31.1% (19/61) and 14.7% (16/109) of Themba hospital and NMAH respondents respectively getting between six and eight correct. Also noted in Table 3 is that 74.0% (279/377) of overall respondents scored below 50%; this was 84.4% (92/109) at NMAH and 66.3% (55/83) at RFH and this difference was statistically significant (p-value = 0.017).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Staff knowledge of policies or practices that were available within their hospitals.

https://doi.org/10.1371/journal.pone.0312340.t003

Further noted in Table 3 and Fig 1, is that the median knowledge scores were 41.7% (25^th percentile = 33.3% and 75^th percentile = 50.0%) in three of the hospitals, and 33.3% (25^th percentile = 25.0% and 75^th percentile = 41.7%) at NMAH. These differences were statistically significant (p-value = 0.002).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Staff knowledge score percentages by years of experience.

https://doi.org/10.1371/journal.pone.0312340.g001

Association between demographic factors and knowledge

Fig 2 shows that only 15.2% (19/125) of those who had not reached five years of experience in that health facility had a knowledge of at least 50% of the available policies or practices, compared to 36.1% for those who had worked for between 21 and 37 years.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Percentage of staff who were affirmative about all protocols/practices in their hospitals by profession.

TH = Themba Hospital; NMAH = Nelson Mandela Academic Hospital; RFH = Rob Ferreira Hospital; SEH = St Elizabeth Hospital.

https://doi.org/10.1371/journal.pone.0312340.g002

Whilst only 27.7% (51/184) of PNs knew of the existence of at least 50% of the present policies or practices, 42.9% (6/14) and 39.2% (20/51) of pharmacists and doctors verified at least half of the policies or practices (Fig 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Percentage of staff who were knowledgeable about all protocols/practices in their hospitals by hospital.

https://doi.org/10.1371/journal.pone.0312340.g003

Table 4 shows that individuals who had more than five years of experience were at least 20% more likely to have gotten 50% or more of the policies or practices correct compared to those with five years or less and these were statistically significant (p<0.05). Whilst PNs were 20% more likely to know at least 50% of the existing policies or practices than other nursing categories (PR = 1.2; 95%CI: 1.1–1.4; p-value = 0.002), medical doctors were 30% more likely to have fared better than the aggregated nursing categories and this was statistically significant (PR = 1.3; 95%CI: 1.0–1.6; p-value = 0.037).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Bivariable associations of knowledge on protocols/practices in their hospitals.

https://doi.org/10.1371/journal.pone.0312340.t004

Discussion

This study determined the knowledge of clinical staff on clinical governance protocols/tools that are available and/or utilised in their hospitals. The research found a majority of participants had sub-optimal knowledge of the protocols or tools that existed in their hospitals. Complaints management and adverse events protocols were two, out of twelve, most known and positively affirmed as available and in use at all facilities. Factors associated with good knowledge of the protocols/tools included more than five years of experience, being a professional nurse compared to other nurses, not working at NMAH as well as being a medical doctor or pharmacist compared to other staff.

In South Africa, the concept of clinical governance has been introduced into different policy frameworks [23, 24]. This study and literature reveal that the drive to improve the quality of healthcare in South Africa has not been lacking in interventions and powerful ideas [25]. However, there is limited literature about evaluating clinical governance implementation and its impact in South Africa. This cross-sectional study is one of the first to determine clinical staff knowledge of clinical governance protocols/tools in selected South African public hospitals. It is reported that creating a sense of ownership, use of educative methods, information sharing, use of procedures and clinical guidelines are some of the effective activities in better implementation of clinical governance [26]. However, the results of this study demonstrate that clinical governance is implemented in these hospitals, but with a concerningly low level of staff knowledge about the concepts of clinical governance. This study therefore validates the importance of ensuring that protocols and guidelines are not only available as a tick box but also ensuring that the mission and vision of hospitals operationalise these protocols and guidelines. In this way, all frontline health workers will be aware of these protocols and guidelines and as such they will have the desired impact.

The results of this study are similarly reported for public hospitals in Tehran [26]. The two mostly known and affirmed protocols; complaints management and adverse events, could be due to more emphasis as these are key priority areas [9, 13]. Which is an emphasis of the continuous intervention for quality improvement and responsibility in all health establishments [27]. These results highlight limited or lack of information sharing on the concept of clinical governance as well as possible non-achievement of desired outcomes to improve the quality of care in their facilities.

Literature [28] has long established that, in order to improve the quality of healthcare, knowledge must be applied. Quality improvement seeks to standardise processes and structures to reduce variation, achieve predictable results, and improve outcomes for patients, health systems, and organisations [29]. Although much has been done over the years to restructure the system and improve the quality of care in health systems, literature reveals that millions of people in South Africa still suffer preventable harm every day [8]. Medical litigation has dramatically increased both in frequency and in the size of the damages [30] reflecting the cost of poor knowledge and application of important protocols or guidelines by frontline workers.

Participants from NMAH had the least knowledge of clinical governance protocols/tools compared to other facilities studied. This continues to cause long delays in the achievement of quality healthcare delivery [31]. Considering that NMAH is an academic quaternary hospital, a better knowledge was expected. The least knowledge of protocols in this academic facility may point to a deeper problem with regards to leadership as previously observed in another publication [32]. Experience and the type of profession were associated with a higher knowledge of the protocols/tools. Despite professional nurses not being fully knowledgeable of existing clinical governance protocols/tools in their facilities, they were more knowledgeable compared to other nursing categories. This could be through repeated exposure and building of a culture that incorporates these protocols into everyday practice, they have become more knowledgeable while other categories are still under supervision. One of the firm foundations for the structure of clinical governance is education and training to raise the awareness of the nature of clinical governance [33, 34]. Therefore, lack of introductory training on clinical governance could have been a significant factor in junior staff having less knowledge about the protocols. Allied (physiotherapist, dieticians etc.) workers were either unsure or didn’t think the protocols and practices were present, this could have been due to the fact that they do not interact with some. Doctors and Pharmacists had better knowledge than other professions. Previous studies have shown that separate meetings and reporting forums for clinical and nursing managers, under the leadership of the Chief Executive Officer (CEO), could be the sole reason for differences between frontline health workers on knowledge of clinical governance tools in their hospital [35].

The design of this study could not allow for a conclusion on the causality of clinical governance protocol/tools knowledge or lack thereof. The use of the quantitative method limited this study in that it typically focused on objective data and did not capture the subjective experiences (deeper insights) of participants about clinical governance implementation at their hospitals. Regardless of the limitations, this study established preliminary evidence for planning of an intervention. Clinical staff often have a first-hand and positive experience of improving quality of care [36]. Therefore, clinical staff in these hospitals need to take the initiative and responsibility in the development and/or review of clinical governance protocols/tools in these hospitals. Managers of the facilities must prioritise clinical governance information sharing with clinical staff. Workshops are recommended to better understand the clinical governance framework, clarify the roles and responsibilities for all players and establish structures of quality improvement in the hospitals. Training workshops must put emphasis on importance of local ownership of the concept of clinical governance, by both management and clinical staff [37]. Communication and clinical governance information sharing amongst clinicians, line managers and policymakers should be at the top of priorities. Junior staff must be introduced to clinical governance in curriculum and during employment induction. The latter must be a practice for all new clinical employees.

Conclusion

The concepts and goals of clinical governance have not been effectively conveyed to frontline health workers in these hospitals. Despite clinical governance implementation, there has been low clinical governance protocols or tools knowledge among the staff. Therefore, providing more effective relevant training workshops, with emphasis on the importance of local ownership of the concept of clinical governance by both management and clinical staff will likely improve knowledge on clinical governance and the quality of care rendered. Clinical Governance must be implemented through comprehensive management support and participation of all staff and health professionals. For junior staff with less than five years of experience it is recommended that they must be introduced to clinical governance tools in their curricula and during induction. Any plans that come out of the intervention must be designed to be sustainable and implementable within current resources and other facilities. Focus areas are to prioritise communication between management and clinical staff and set up measures for the impact of the intervention.

Supporting information

Acknowledgments

The authors would like to thank the South African Medical Research Council, Division of Research Capacity Development, the officials of the participating hospitals and provinces for their support and assistance. We would also like to thank our research assistant, especially Mr Siphosethu Xinwa for making this research possible.