This project aimed to improve the documentation of nursing assessments in the patient record as a step toward improving nursing practise at the hospital.

CCHU: CCHU is a specialist paediatric neurosurgical teaching hospital, located in Mbale, Eastern Uganda. The hospital's focus is treatment of children with hydrocephalus, neural tube defects (including spina bifida), and epilepsy and brain tumours.

During the study period (2008–2012), the hospital had 40 beds (30 ward beds and 10 Intensive Care Unit [ICU] beds). There were, on average, 1230 admissions each year; 1180 of these children were being seen for the first time.

At CCHU the nursing service is led by a nurse manager and an assistant-nurse manager. A charge nurse on each shift is responsible the day-to-day running of each of the ward, ICU and operating room or theatre. The nursing care activities are organised under a total patient care delivery model in which the primary care nurse is responsible for all the care of a group of patients throughout the shift. The hospital employed 34 primary care nurses, the majority of whom were enrolled nurses. Enrolled nurses are registered with the Uganda Nurses and Midwives Council after a three year training programme in general nursing science or comprehensive (nursing & midwifery) nursing science. This level of training requires an Ordinary-level certificate (11 years of schooling) for entry into the training programme. Diploma and baccalaureate nursing training programmes require an Advanced-level certificate (13 years) and graduates are registered nurses.

Uganda: The economic cost of treating children with neurosurgical disease is substantial in developed and low resource countries alike (Rabiu & Adeleye 2013). Location in the developing world makes achieving quality health care delivery a difficult goal for a variety of reasons, not least of which is the limited availability of resources. Competent nursing as a resource is fundamental if quality outcomes for patients and hospitals are to be realised. In the authors’ experiences of teaching in nursing colleges in Uganda (i.e. non-baccalaureate training institutions), critical thinking (including models like the scientific nursing process), clinical reasoning and evidence-based practise are infrequently emphasised in nursing education. The principle approach to teaching and in clinical settings seems to be task-based rather than patient-centred. Recent evidence shows that task-based care is associated with increased risk for adverse events (Dubois et al. 2013:113).

Baccalaureate nurse training in Uganda started in 1993 (Klopper & Uys 2013). Both nursing service leaders were bachelor-prepared nurses who had a sound knowledge base in scientific processes and patient-centred care. The nursing leadership was able to identify the first step in changing established practise (from a task-based model to patient-centred practise) as the need to improve the nurses’ knowledge and skills in assessment and documentation. The long-term goal was to achieve professional nursing practise that met the needs of the high acuity, neurosurgical paediatric patients and their families.

Documentation is reported to take up to 50% of nurses’ time per shift (Gugerty et al. 2007:6). It serves a number of important functions, including communication amongst healthcare workers for continuity of care. Poor communication, in a broader sense, is known to contribute significantly to the occurrence of adverse events in healthcare (The Joint Commission 2012) and is therefore an important target of initiatives to improve patient safety. Additionally, documentation is important for education, research, quality assurance and for reimbursement by third party claimants (DeLaune & Ladner 2011:216). According to Wilson et al. (2012), nursing documentation can serve as an indicator of quality nursing practise. Additionally Collins et al. (2013) demonstrated that nursing documentation could be used to predict mortality.

According to Jefferies, Johnson and Griffiths (2010:119), quality nursing documentation meets seven criteria: (1) patient-centred, (2) contains the actual work of nursing, (3) reflects the nurses’ clinical judgment, (4) is presented in a logical sequence, (5) is written in real time, (6) records variances in care and (7) fulfils legal requirements. Prideaux (2011) argues that the quality of nursing documentation in many countries remains poor due to numerous reasons which different authors articulated. A Dutch study found inadequate documentation of important aspects of assessment and other related nursing care, including inaccuracy of documentation (Paans et al. 2010). A Swedish study reported incongruence between what is documented and actual physical patient status (Gunningberg & Ehrenberg 2004), and limited documentation of the work of nurses in an Australian study was documented by Jefferies, Johnson and Nicholls (2011). These inadequacies could be related to challenges in two areas: the nurses’ individual characteristics and work environments (Cheevakasemsook et al. 2006; Taylor 2003).

Nurses’ perceptions and attitudes towards documentation impact the quality of how and what they document. Studies in various settings found that whilst nurses consider documentation important for nursing professionalisation, they consider it a burdensome secondary task that takes nurses away from direct patient care (Blair & Smith 2012; Björvell, Wredling & Thorell-Ekstrand 2003b). Also, as nurses consider increasing liability for their practise, their documentation may be negatively affected (Brown 2013; Cartwright-Vanzant 2010). Furthermore, knowledge of the nursing process, which forms the basis for nursing documentation, has been found to be inadequate (Ofi & Sowunmi 2012; Allen 1998).

The workplace environment can contribute to poor documentation. Heavy workloads, laborious documentation forms, fragmentary language (i.e. documentation language that is not understood beyond the local context), inadequate resources and hospital culture all impact the quality of nurses’ documentation (Prideaux 2011; Jefferies, Johnson & Nicholls 2011; Taylor 2003; Björvell, Wredling & Thorell-Ekstrand 2003b).

Given the significance of nursing documentation and the reality of poor documentation practises, it is not surprising that there have been resolute calls and subsequent efforts to improve its quality (Whitcomb et al. 2013).

Training nurses to improve knowledge, skills and documentation practises has been a widely used strategy. For example, Jefferies et al. (2012) used a writing coach programme to improve documentation quality. Training has been augmented by using written practise standards (Considine, Potter & Jenkins 2006); standardised nursing records, or a nursing model (Tornvall, Wahren & Wilhelmsson 2009; Björvell, Wredling & Thorell-Ekstrand 2003a).

Data from patient records revealed poor documentation of nursing assessments. Informal interviews of nursing leadership (key informants) identified inadequate training as a likely explanation for substandard documentation.

In October 2008 a five-day training workshop for all nurses employed at the hospital was planned and conducted by a multi-disciplinary team involving the hospital anaesthesiologist, medical officers, nurse managers and human resources staff (see Figure 1, Phase 1). The training comprised topics on assessment and professional nursing practise. Trainees had access to preceptor and mentorship support post-training.

One month following the training, the researchers compared all records of patients admitted and discharged during a one week period.

Whilst training participants displayed improved assessment techniques, the chart audit showed no observable change in documentation practises. Comparison of the 2008 post-training documentation practises and content showed they were very similar to records made by different nursing personnel in 2001–2003. Therefore, training alone was insufficient to change practise.

Reflection on these findings pointed to a cultural component of documentation practise that would need to be addressed to achieve observable change in documentation of nursing assessments. The culture was defined in relation to the values, attitudes, competencies and patterns of behaviour (Feng, Bobay & Weiss 2008) of the nurses at the hospital.

In Phase 1, the researchers identified a cultural component of nursing documentation practise that was a barrier to change and conducted a literature review to gain further understanding of the concept of culture in healthcare. Interviews with nursing staff, and visiting nursing students and professors from the United States provided additional perspective.

Findings guided modifications to the content and organisation of the hospital's traditional two-month orientation programme for newly hired nursing staff (Table 1). The modifications aimed to teach a new patient-centred way of thinking and doing, to consolidate the new preferred knowledge and documentation practises, and to create a critical mass of staff that would help drive the efforts to change the culture and sustain the practise change and improvement process.

Topics from the 5-day workshop on documentation were incorporated into the orientation to address identified gaps. The first month of orientation was changed to exclusively classroom-based training with only limited ward time so that the new nurses learned the preferred knowledge and practises prior to exposure to the targeted undesirable practises and cultural norms. The second month of orientation included a mentorship period that socialised the new nurses to the ward, operating room or theatre and ICU environments. During this second month of orientation, the new nurses shadowed full-time employees and were oriented to other hospital departments (e.g. radiology, laboratory and physiotherapy).

Further, the researchers developed a new form to facilitate complete documentation of nursing assessments on admission and discharge. The form prompted documentation of a comprehensive nursing history and physical examination aligned with processes taught during orientation.

Charge nurses on the wards reviewed and commented on the redesigned assessment documentation form.

This second cycle of action research focused on initiating the process of culture change in two areas: professional nursing development – orientation, and systems changes – the redesign of the assessment documentation forms.

The charge nurses indicated that the first draft of the revised documentation form – a ruled paper with headings for complete health assessment – was unduly laborious. The form required much writing and relied on nurses’ recall of what to document. Therefore, the form needed modification before it could be introduced to the wards.

After implementation of the modified orientation programme, but before implementation of the new documentation forms, chart audits revealed improvement in newly hired nurses’ documentation at admission, and on a shift-by-shift basis. However, this change was not sustained; in spite of continual feedback and training, the documentation content and practise of new nursing staff reverted to the established inadequate documentation style.

Reflecting on this, the researchers concluded that the power of the culture was too strong to sustain improvement with new nurses quickly taking on the prevalent practises. It was concluded that the form would need to be easier to use in order to support adoption of the new form and documentation practises.

Phase 2 initiated the process of culture change. The purpose of Phase 3 was to build on the changes that, though not sustained, had been visible. A literature review on culture, quality improvement and change management focused reflection on systems changes as drivers of changes in practise. Necessary systems changes included: (1) redesign testing and implementation of the admission assessment form, (2) change in employment policy, and (3) new focus on creating a healthy working environment based on the American Association of Critical-Care Nurses’ (AACCN) standards (2005);

Form redesign: Beginning in June 2010, the corresponding author and a group of nurses in orientation training redesigned the admission assessment form with review and commentary by charge nurses. The modified form was pilot tested with the ward nursing staff during August 2010. Results from the pilot guided further modification and retesting during October 2010. Nurses began using the new Nursing Admission and Discharge Record form in January 2011 (see Appendix 1).

Learning from the previous phase, the researchers planned to refine the documentation form(s) and modify other systems factors (appropriate staffing and healthy work environment), to achieve sustainable change in documentation practise:

The nursing admission and discharge record: This form was redesigned to make it simple to use, requiring standardised content, limited need for recall and minimal writing, whilst still enabling the capture of all information needed for the comprehensive nursing assessment. The form is a 3-page admission and discharge record requiring check off of items needed for a comprehensive assessment of a child with neurological disability in the local context (Appendix 1). It requires documentation in a standardised format that reflects a deliberate thinking process and problem-solving approach, based on the American Heart Association's Paediatric Advanced Life Support assessment (Kleinman et al. 2010). New nurses were trained, audited and supervised using this assessment approach and it formed the basis for continuing education efforts.

The CCHU nurses’ experiences of providing care to children with neurological illness informed content revisions along with new evidence regarding the follow-up of children with neurological disability. According to Warf, Wright and Kulkarni (2011), mortality of children with spina bifida is more strongly related to their home and/or community situation than in-hospital factors. Consequently, a section to gather information about the child's and family's home and community environments, especially healthcare services and social support systems, was incorporated into the admissions form. The form includes a focus on family composition and the parents’ relationships with in-laws who, according to information gathered from the parents, seemed to have a significant influence on how family decisions are made, including decisions on healthcare seeking behaviour and follow-up. The researchers anticipated that collecting this data would alert the nurses to child and family needs and inform an individualised plan that would facilitate adherence to the follow-up schedule and positively influence patient outcomes. Further, the revised documentation form assesses spirituality and belief systems. Furthermore, screening on admission for sepsis and shock was included based on anecdotal evidence in this setting that suggested that these conditions were prevalent and often missed:

Skill mix of nurses: Up to 2010, the nursing skill mix at the hospital comprised both enrolled and RNs, with a predominance of enrolled nurses (see Table 2). Considering the limited resources, along with the need for sustained improved outcomes, nursing leadership decided to employ only RNs. This decision was supported by research evidence showing that increasing the proportion of staff nurses with higher levels of education ultimately led to improved outcomes (Aiken et al. 2003). Further, the model of total patient care delivery used at the hospital requires a widened scope of practise, strong critical thinking and clinical reasoning skills (Tiedeman & Lookinland 2004) more characteristic of RNs. Retention of RNs was also better than the retention of enrolled nurses. Consequently, in 2013 there was a higher proportion of RNs providing direct care at the hospital: 88% compared to 31% in 2008 (Table 2).