Undertaking general practice quality improvement to improve cancer screening - a thematic analysis of provider experiences

Background Cancer is a major cause of illness and death, and its incidence and mortality can be reduced through effective screening. In order to improve below target screening rates in one region of Australia, the local Primary Health Network supported local general practices to implement a range of quality improvement initiatives. Methods We used a qualitative approach and interviewed 18 general practice staff and five Primary Health Network staff and contractors to understand their experiences with these quality improvement initiatives. Results In a thematic analysis, we identified four key themes related to program set-up and implementation; patient and community education and promotion; engaging patients and communities in screening; and general practice enhancement. Program roles were clear and understood, and the program received strong oversight and support. Practice staff felt supported and motivated. Information Technology was a challenge for many practices often requiring tailored assistance. Education provided by practices facilitated patient empowerment but practice staff noted difficulties engaging patients in screening. Practices were enhanced though strong leadership and teamwork and practice learning activities. Conclusions The tailored evidence-based quality improvement initiatives were considered effective in supporting general practices to increase their cancer screening. Key facilitators reported by participants included use of Plan-Do-Study-Act cycles, enhanced data entry and audit capacity, effective recall and reminder systems and maintaining staff motivation. Supplementary Information The online version contains supplementary material available at 10.1186/s12875-021-01581-y.


What is known about the topic?
Despite the recognised benefits of quality improvement in general practices there have been no evaluations of programs applying a broad range of practice-based initiatives specifically to cancer screening.

What does this paper add?
The present paper provides evidence on how implementing a tailored combination of quality improvement initiatives in general practices can rapidly improve cancer screening participation.

Background
Cancer is the second leading cause of death globally [1] and the leading contributor to disease burden in Australia [2]. As a major cause of illness in Australia, cancer has a substantial social and economic impact on individuals, families and the community [3]. Cancer screening programs aim to reduce cancer incidence and mortality [4] and national screening programs are available in Australia to detect breast, bowel and cervical cancer in targeted population groups [5]. Screening for cervical cancer is conducted in general practices, and they inform and encourage patients to enhance participation in national breast and bowel cancer screening programs [6]. BreastScreen Australia is a joint initiative of the Australian and state and territory governments with screening conducted by accredited screening services [7]. The Australian government funded National Bowel Cancer Screening Program is based on direct mailing of faecal occult blood detection kits to those eligible for screening [8]. Several studies have shown that GP endorsement of bowel cancer screening is an effective method of increasing participation [9]. Primary care endorsement of breast and cervical cancer screening has been similarly effective in increasing participation [10].
Primary Health Networks (PHNs) are regional organisations funded by the Australian Government to improve efficiency, effectiveness and coordination of primary health services in their region [11]. These organisations support general practices to undertake quality improvement including through use of practice data [12]. Improving participation in cancer screening in the Nepean Blue Mountains population is a priority for Nepean Blue Mountains Primary Health Network (NBMPHN), which supports general practices in four local Government areas to the west of Sydney, where participation rates across Australia's three National cancer screening programs (bowel, breast, cervical) are below New South Wales (NSW) State averages [13][14][15][16]. Cervical screening rates in this region in 2014-15 were 53.3% (compared to the State average of 56%) and breast cancer screening rate was 46.2% (compared with 51.6%), with bowel cancer screening rate 33.3% (compared with 35.1%).

Nepean Blue Mountains primary Health network Cancer screening program
With funding from the Cancer Institute NSW, from 2016 to 2018, the NBMPHN worked with communities to raise awareness of cancer screening and with general practices to improve screening and detection of breast, cervical and bowel cancer through evidencebased strategies tailored according to individual practice needs [17]. This included training practice staff in development of screening registers and patient recall systems, improvement of data entry, and use of audits in Plan-Do-Study-Act (PDSA) cycles. The PDSA cycles provide for iterative testing of changes to improve quality of healthcare and healthcare systems [18]. Health promotional resources were also provided for patients as well as on-line clinical and referral guidelines for general practitioners (HealthPathways). An Aboriginal liaison and a community educator were commissioned by the NBMPHN to engage patients in cancer screening (Table 1).

Research aims
As part of an evaluation of the NBMPHN Cancer Screening Program (NBMPHN -CSP), we aimed to explore how general practice and PHN staff experienced the general practice based quality improvement initiatives described above. Our findings are likely to inform other preventive health initiatives undertaken in partnership with general practices. Consumer perspectives including of community based educational activities, will be reported in a forthcoming manuscript.

Evaluation scope and oversight
The researchers (CM, ST, JR) from Western Sydney University developed a program logic model (PLM) in consultation with the program advisory committee, and from reviewing the literature and program documents, to guide the evaluation activities and provide a comprehensive framework for future cancer screening evaluations by the NBMPHN (Additional file 1 provides a summarised PLM). A PLM provides the capacity to extensively investigate all components of a program using multiple data collection methods and re-implement the framework for future evaluations [19].
To evaluate changes to bowel, breast and cervical cancer screening rates across its region following the strategies described above, a Western Sydney University team of qualitative researchers (CM, ST, JR) were commissioned by the NBMPHN. We aimed to explore facilitators and barriers identified by participants that could inform ongoing implementation of the quality improvement strategies.
The NBMPHN established a program advisory committee consisting of program management, and academic, clinical and consumer representatives, oversee the implementation of the quality improvements and guide the evaluation team.

Setting
The Nepean Blue Mountains region west of Sydney, comprises four local Government areas including urban and semi-rural areas which cover almost 9179 km 2 [21]. Transport availability, distances to services especially for outlying areas, and costs are dominant issues for the region [20]. The NBM region is aging at a faster rate compared to the rest of NSW with the increase in older persons as a proportion of the population, 5.13% compared to 3.3% across NSW between 2011 and 2026 [21].

Study design
We conducted a qualitative evaluation of the NBMPHN-CSP using semi-structured interviews aligned with the PLM. We used the COREQ criteria as a guide for reporting our research [22].

Participant recruitment
The program advisory committee identified and recruited a purposive sample of general practitioners, practice nurses and practice managers engaging in the NBMPHN-CSP. Potential participants were contacted using a letter of invitation and information/consent form approved by an ethics committee, and participants contacted us to schedule interviews. We stopped recruitment when we approached our target of 20 general practice participants and when adequate representation by practice staff type and locality was achieved. This predictive sampling is supported by research that notes the first five or six interviews produce the majority of new information in a data set [23]. We also sampled participants who were less positive about the program by identifying those practices that were reported by PHN staff to be less well engaged with the quality improvement interventions. The researchers contacted a small number of PHN program staff directly, including an Aboriginal liaison and community educator for interview as their perspectives of the program implementation were likely to provide important insights. The PHN staff included program officers and management who were familiar with their local area and experienced in working with general practices . No participants withdrew their consent.

Data collection and analysis
In consultation with the program advisory committee, we developed a semi-structured interview schedule aligned with the relevant indicators of the PLM. We explored participant experiences with each of the individual quality improvement strategies, and how they were implemented, as well as any facilitators and barriers encountered. Questions also included how practices were oriented to the program, the supports provided, and outcomes at a practice level. All interviews were approximately 30-40 min in duration, audio-recorded and transcribed, and conducted one-on-one by two researchers (ST, CM), either face to face in private offices or by telephone as preferred by participants. All participants were given the opportunity to review their transcripts. We piloted the first five interviews to ensure the schedule captured the required data. Three research team members (CM, ST, JR) reviewed these first interview transcripts. The interview guide underwent further minor revision during data collection, with new questions and prompts added to explore emerging areas of interest. This process was informed by ongoing analysis of each interview as it was transcribed and by input from the program advisory committee (Additional file 2).
We conducted an inductive thematic analysis to interpret the experiences and perspectives of participants with the NBMPHN-CSP. This approach allows patterns and meanings to be captured from qualitative datasets [24]. We used a reflexive and collaborative approach to coding designed to develop a richer more nuanced reading of the data [25]. Research team members (CM, ST, JR) each coded three -four of the first eight interviews to identify patterns in the transcripts. We then agreed on an initial coding frame and coded the remaining interviews and consulted together to check and refine the emerging analysis and consider any differences in interpretation. At a final workshop, the researchers (CM, ST, JR) reviewed all interviews and agreed that saturation of codes had been achieved. The final thematic structure was also agreed to clearly and comprehensively describe our analysis (Additional file 3). We used N-Vivo 11 ® software to help organise the interview data.

Results
We interviewed 23 participants over a four-month period from December 2017 including general practitioners, practice nurses and managers, and program support staff from the NBMPHN. Practice staff were drawn from the four local Government areas of Nepean Blue Mountains ( Table 2).
We identified four key themes related to program setup and implementation; patient and community education and promotion; engaging patients and communities in screening; and general practice enhancement. These themes and the related subthemes are described in Table 3 and detailed below.

Setting up and implementing the Cancer screening program
Interviewees noted that most program and committee staff, and contractors, had a good understanding of their roles, and expectations were made clear. Most staff felt well supported and knew where they could seek assistance. A strong governance structure was noted with consumer and clinical representation on the program advisory committee. Staff described the program as evidence-based, and similar to other well-evidenced programs. Funding was mostly considered adequate with practice payments described as helpful and an incentive to join the program, even if not covering additional staff time. Distribution of payments to individual GPs rather  than to the practice as a whole, was raised as a concern by some interviewees.

Patient and community education and promotion
Participants described practices providing patient education on screening through brochures and posters in waiting rooms, practice websites, at regular health promotion days and opportunistically during consultations. Some practice staff saw promoting screening as a way to improve knowledge and attitudes, including in the wider community. Screening education was regarded as empowering consumers and it was also encouraging for practice staff to see patients engage in cancer screening.
The last three results in some women's files is their mammogram, their FOBT, and a cervical screening, so they seem to be doing it simultaneously, they're like, "okay well I'm on the bandwagon I might as well get it all done now". PN 1

Engaging patients and communities in screening
Cancer screening became a practice priority promoted through team meetings and informal conversations. Systems were developed or improved such as practice registers, recall and reminder systems and practice data collection and audit. Practice staff noted that regular use of these systems encouraged patient awareness and participation. Interviewees noted that the PHN provided learning activities and responsive support throughout the program. Information was available through websites, face to face learning and through a range of resources such as screening Health Pathways and "cheat" sheets for practice staff working with IT. Practice staff described how training improved their efficiency. They became more aware of screening rates and proficient with data entry and cleansing. The PHN considered the support they were providing to general practices as crucial in sustaining improvements achieved. Practice staff expressed commitment to continue quality improvement initiatives but some also recommended ongoing PHN support to maintain focus on cancer screening. Most respondents thought data collection and analysis should be performed by the PHN.

Discussion
Our findings report the experiences of those engaged in the NBMPHN Cancer Screening Program. Analysis of the interview transcripts revealed four overarching themes: setting up and implementing the Cancer Screening Program; patient and community education and promotion; engaging patients and communities in screening; and general practice enhancement. As noted by other research using multimodal quality improvement strategies in primary care [26], our qualitative research findings support the effectiveness of such activities in engaging practice staff and patients. Facilitators and barriers identified included the importance of strong oversight and governance, as well as collaborative relationships and organisational support to overcome problems with information technology and enhance the use of data. Maintaining motivation with quality improvement was also regarded as crucial. These are discussed below with reference to the literature. Support for quality improvement in primary care can enhance uptake of evidence-based practices and improve patient care, however quality improvement can be difficult to implement and sustain [27]. Strong program oversight and direction are essential in supporting quality improvement initiatives [28,29]. The PHN established clear governance structures for the program, including appropriate cultural and clinical representation such as an Aboriginal community member and general practitioners. This supported effective engagement with the community and general practices.
Quality improvement also requires collaboration, with trusting and respectful relationships critical for adoption of evidence-based healthcare improvements [30]. Key to improving screening rates in Nepean Blue Mountains was the tailored support provided to general practices. Consistent with other research, our evaluation noted the many competing priorities in general practices, and limitations in information technology skills [31]. General practice staff described problems with software programs and their communication with external providers such as pathology, resulting in challenges with data availability, entry and extraction. They valued PHN staff who provided individualised practice support to address these challenges. Tailored, hands on support by PHN staff who have longstanding relationships with the practice are critical for quality improvement in general practice, however, practice staff also need to engage in the support and training provided.
Meso-level organisations such as PHNs have an important role in facilitating data measurement for quality improvement, and providing incentives and professional education [28]. The PHN assisted practices to establish and refine screening registers and recall and reminder systems, conducting periodic audits, as well as IT updates and troubleshooting. The PHN also provided training in planning and implementing PDSA cycles, and directed practices to other resources such as Health Pathways for cancer screening. Where similar support has been provided for colorectal cancer screening, especially using a team-based approach, improvements have been achieved in staff engagement and practice efficiency [32]. Training and provision of ongoing access to resources will ensure the maintenance of quality improvement.
Maintaining motivation and engagement is also critical in sustaining quality improvement [33]. Quality improvement has been described as a "team sport" where collaborating team members support and motivate each other toward common goals [34]. Our participants considered regular team meetings with engaged leaders to be highly motivating however, as noted by others [30], some of our participants reported attitudinal barriers and practice leads who were not engaged in the program. They expressed frustration and noted difficulties engaging with quality improvement when this commitment and guidance was lacking. Disengagement was said to occur when there was poor team communication and collaboration, all of which affected staff confidence. Efforts to build a team commitment to quality improvement and improving communication within teams are required when implementing quality improvement.
Most of the participating general practices described a keen sense of engagement and motivation. They received financial incentives and CPD credit for their participation but, consistent with other research, most did not consider these extrinsic rewards as key motivators [35,36]. Instead, many described the intrinsic rewards of improved skills and efficiency, progress demonstrated by benchmarking, and especially improvements in patient care through increased cancer screening activity. They also described support from the PHN as motivating and key to sustaining practice improvements achieved. Strong ongoing collaboration with and support from PHNs are essential in maintaining motivation and engagement in quality improvement [27,29]. To sustain quality improvement initiatives, consideration needs to be given to continuing support and ongoing motivation.

Strengths and limitations
Our in-depth qualitative evaluation with a range of stakeholders provided valuable insights that can inform implementation of other quality improvement initiatives in general practice including beyond cancer screening.
Although we interviewed staff from practices struggling with the program, a limitation of the research is that we did not interview staff from general practices that chose not to participate in the quality improvement initiatives. This may have provided further insight into barriers and needs which if addressed could enhance general practice engagement with such programs.

Conclusions
Primary Health Network provision of a range of evidence-based, tailored, quality improvement supports was effective in supporting general practices to improve cancer screening. Key facilitators reported by participants in our research included the need for strong program oversight, continued individualised support from PHN staff, ongoing access to resources and training, team commitment through improved communication, and continuing strategies that maintain staff motivation.