Implementation of a Specialized Prostate Cancer Clinic in a Resource Limited Setting: A Mixed Methods, Retrospective, Process Evaluation study.

Prostate cancer represents a critical oncology challenge in sub-Saharan Africa, where health system constraints have created a pronounced evidence-to-practice gap. Recent epidemiological data indicate that prostate cancer has become the leading incident malignancy among men in 40 African countries [1].  Nationally, in Zambia, prostate cancer (11.2%) is the leading incident male cancer and the third all sexes cancer after cervical cancer (23%), Kaposi’s Sarcoma (16%) [2]. This disease burden occurs within a context of significant cancer care resource limitations, where less than 10% of Zambian health facilities currently provide prostate cancer services [2].

The challenges in prostate cancer care persist despite the existence of evidence-based diagnostic and therapeutic protocols. This disparity highlights the critical need for implementation science approaches to clinical care. As defined by Bauer et al [3] , implementation science systematically examines methods to promote the adoption of research findings into routine practice, focusing specifically on overcoming barriers at provider, clinic, and healthcare system levels [4].  This discipline differs fundamentally from quality improvement initiatives through its emphasis on developing generalizable knowledge and its application of rigorous evaluation frameworks [2]. Additionally, implementation science seeks to develop generalizable knowledge that extends beyond the specific system being studied. This approach typically involves three key types of evaluation: process, formative, and summative evaluation [1, 4].

The current study applies implementation science methodology through the Consolidated Framework for Implementation Research (CFIR) [5] combined with a process evaluation approach to assess a prostate cancer service improvement initiative at Ndola Teaching Hospital. Historically, prostate cancer patients have been managed across multiple clinics and health centers around Zambia [6], with little to no specialist input, this has led to fragmented care, redundant efforts and inefficiencies. The study targeted three key areas identified through preliminary process mapping; Service consolidation through dedicated prostate cancer clinics, enhanced patient identification systems and standardized record integration.

This study aimed to document and analyze existing prostate cancer services at Ndola Teaching Hospital, implement an improved service model, and evaluate its effectiveness. With prostate cancer mortality in Africa projected to increase from 47,000 in 2020 to 100,000 by 2040 [7] such implementation-focused research represents an urgent priority for global surgery equity.

1. Study setting & Population.

The study was conducted at the urology outpatient department of Ndola Teaching Hospital, a tertiary referral centre serving Zambia’s Copperbelt Province, with a catchment area serving a population of approximately ~2 million people. The clinic operates twice weekly, staffed by 4 urologists per clinic day, three nurses, and two clerks . The study population was as follows.

1. Outpatient department specialist clinic clerks: Registry staff responsible for patient registration and record keeping during the study period.
2. Outpatient department specialist clinic Nurses: Nursing staff stationed in the OPD specialist clinics specifically those allocated to urology during the study period.
3. Urology doctors: the doctors responsible for outpatient urology services during the study period

2. Data collection:

Data was collected during the study period from 1st October 2023 to 31st December 2024. October 2024 was the start point marking the decision to implement the clinic, while December 2024 was the end point as this is the time annual OPD registers are closed for a given year.

Quantitative data:

1. Diagnostic records : Monthly counts of PCa vs. BPH diagnoses from OPD registers during the study period
2. Time to diagnosis: Number of Days from initial urology visit to visit for confirmation of diagnosis, collected from OPD records during the study period

Qualitative data:

1.Process mapping: Direct observation of clinic workflows.

2. Staff interviews: anonymous Online questionnaires before and after implementation of the intervention.

Inclusion criteria

1. All urology OPD clinic diagnoses recorded during the study period
2. All urology OPD clinic patient visit records entered during the study period
3. All Urology OPD staff stationed in the clinic during the study period

Exclusion criteria

1. Incomplete diagnoses’ record entries.
2. Incomplete patients visit records
3. Temporary Urology clinic staff not permanently stationed in the Urology OPD

 iii. Intervention

The new prostate cancer clinic was implemented within the existing Urology Outpatient (OPD) service and focused on three focus areas: service consolidation, patient file identification and record integration.

1. Service Consolidation:
   • A dedicated prostate cancer clinic, with an allocated urologist, and conducted at specific times, within the Urology OPD was instituted to reduce patient overlap with other urology services.
2. Patient file Identification System:
   • A coding system (file labels) was started to identify prostate cancer patient files.
   • Registration clerks were involved in the coding and prioritization of prostate cancer files to ensure accurate and file management.
   • Urology nurses were involved  in patient file recognition and appropriate allocation to prostate cancer clinics
3. Record Integration:
   • A simple summary sheet was designed and attached to every prostate cancer file; this consolidated diagnosis information, staging and current treatment onto a single pageant was attached at the beginning of the file.

1. Outcome Measures

The following outcome measures were assessed to evaluate the impact of the clinic service improvement during the study period:

• Clinic efficiency: through analysis of clinic process flow
• Accuracy of diagnosis records: The accuracy of the initial versus post intervention recording of prostate cancer diagnoses in the outpatient register
• Time to final diagnosis: the time from first attendance in urology clinic to time of prostate cancer diagnosis

v. Data Analysis

Continuous variables (time-to-diagnosis) were  reported as means ± SD, while categorical variables (diagnosis counts) were reported as frequencies. Pre and post intervention data comparisons were made using paired t-tests; while correlations (PCa Vs BPH recorded diagnoses)were assessed via Pearson’s r (SPSS v23). Qualitative data (processes and workflow) were summarized and presented in flow diagrams using the PlantUML software.

vi.  Institutional Permissions and Ethical Considerations

Permission to conduct the study was obtained from the hospital management. The intervention was aimed at improvement in service delivery within the existing clinical framework. All patients were treated by specialist-led, qualified urological teams, ensuring high standards of care. Any and all patient related data were de-identified while staff surveys were conducted via anonymized online surveys.

1. THE CLINIC PROCESS FLOW

Pre- intervention

Through direct observation between October 2023- January 2024 and combined staff  questionnaires, the existing process was mapped out as below ( Fig 1). The existing process began with a patient attending the OPD Specialist clinic under urology either through a direct referral from another specialist clinic ( OPD-to-OPD referral), referral from the Casualty services ( Casualty-to-OPD referral) or as a follow up visit after a urology admission ( Internal urology referral). The clerical staff are responsible for retrieving the patient’s record and handing the folder to the urology nurse, who then distributes these folders among the urology doctors. Simultaneously, the patient must ensure that he is aware of which doctor his file has been assigned to and sits in the correct waiting area. There is no differentiation of conditions thus, the prostate cancer patients are combined with the other general urology patients. Furthermore, patients are allocated to the varying clinic days thus if one happens to attend clinic on a different day they are required to instead, return on the correct clinic day. The actual consultation required the doctor seeing the patient on that particular day to familiarize themselves with the entire history and treatment progression. This process was repeated on every subsequent patient visit.

 Post- intervention

Through direct observation and combined staff interviews the new process was mapped out below (Fig 2). Post intervention, the referral process remained unchanged; with a patient attending the OPD Specialist clinic under urology through the standard hospital referral processes. The clerical staff remained responsible for retrieving the patient’s record and identification- this was improved by file labels. The nurse remained responsible for directing the prostate cancer patients to the designated Doctor. In the improved service, a particular doctor was responsible for the prostate cancer service per clinic day, furthermore, the consultation process was now improved by the record consolidation sheet (Fig 2).

2. ACCURACY OF DIAGNOSIS RECORDS

Prostate Cancer Diagnoses:

The recorded diagnoses of prostate cancer increased over the study period. Prostate cancer diagnoses before January 2024, ranged from 5 -12 cases per month. From January 2024, there is a rise in the recording of prostate cancer diagnoses (Table 1). There is a statistically significant difference in the average number of both BPH and PCa diagnoses before and after the initiation of the prostate cancer clinic (p=0.002).  The distribution reveals a negative correlation between the two diagnoses, with a downward tendency in the BPH cases versus an upward trend in the Prostate cancer diagnoses (Fig 4).

BPH Diagnoses: The number of recorded BPH diagnoses shows a decline between October 2023 to August 2024. The highest number of BPH diagnoses was recorded in October 2023 and decreases each month, reaching 43 in December  2024 (Fig 3).

3. TIME TO FINAL DIAGNOSIS

In the study period, the mean time to diagnosis, reduced from an average of 83 days to 23 days in 2024. This represents an approximately 60% reduction in time to diagnosis. There is a significant reduction in the standard deviation in the time to diagnosis, from 65 before study period to 15 at the end of the study period (Table 2). Due to insufficient data from before the study period, comparative analysis was not conducted. However, the data suggest that the difference in time to diagnosis is indeed notable.

This study demonstrates that a dedicated prostate cancer (PCa) clinic, implemented through a structured process evaluation framework, significantly improved diagnostic recording accuracy, clinic efficiency, and time-to-diagnosis in a resource-constrained Zambian setting. The piloting of a specialized prostate cancer clinic was based on data supporting early contact with appropriate diagnostic service resulting in improved outcomes [8]. The converse is also true as delays in referral to tertiary centers have been shown to impact the clinical outcomes and disease severity for prostate cancer [9], Zambia has a significant PCa burden and part of the national cancer strategic plan includes improving prostate cancer services (2). While the evidence supporting the positive impacts and clear need for prostate cancer services exist [10], clinical service in the country does not reflect this [6]. From the results, the study has shown that the implementation of a specialized prostate cancer clinic has a significant and documentable impact. The use of the CFIR tool in the methodology presented a unique approach to this clinical improvement project.

Impact on diagnostic records accuracy

The results of the study show that involving non clinical staff in the implementation of the clinic lead to improved recording of accurate diagnoses. Prior to the intervention, clerical staff frequently misclassified PCa as benign prostatic hyperplasia (BPH) in outpatient registers, echoing findings from Lombe et al [6] on data gaps in Zambian cancer care. Post-intervention, diagnostic recording accuracy improved markedly- PCa recordings increased from 7.33 to 14.58/month, p=0.002, underscoring the pivotal role of non clinical staff in success of this implementation project. This finding mirrors CFIR’s emphasis on “individual characteristics” of a system that may affect the implementation process; by adding labels to files and involving the registry clerks in prioritization of PCa cases, we addressed a systemic bottleneck often overlooked in clinical research.

Impact on the time to diagnosis

Secondly, time-to-diagnosis which we used as a proxy for clinic efficiency revealed  a 72% reduction in time-to-diagnosis from an average of 83 to 23 days. This reflects an improved and streamlined workflow, similar to outcomes reported in high-resource “one-stop” PCa clinics [8]. In our context, this was achieved by consolidating records into standardized summary sheets and designating clinic days. Notably, the reduced standard deviation (65 to 15 days) suggests greater equity in access, a critical metric for low-income settings where delays disproportionately affect many patients. This is in agreement with findings shows that access to specialist urology services in a tertiary setting improves clinical outcomes [11]. Furthermore,  there is growing recognition that improvement in patient flow and the capacity to provide patients with timely access to appropriate specialist services is critical in the whole patient journey [12]. This is particularly relevant in Ndola, where observational data from clinical audits of the Urology department at the Ndola teaching hospital reveal that the average life span from diagnosis to death stands at approximately 18 months in 40-60% of cases at the facility.

Cost effectiveness and replicability

The clinic’s integration into routine urology service, without additional staffing or infrastructure, enhanced its adoptability, a core CFIR domain(13). This aligns with Lombe et al [6] call for holistic approaches to cancer care in Zambia and contrasts with other initiatives that may often fail to transition beyond pilot phases. The study directly supports Zambia’s National Cancer Control Strategic Plan (2022–2026), which aims to reduce PCa mortality by 15% through improved service access while our use of CFIR bridged the gap between theoretical frameworks and on-the-ground realities. Intervention Adaptability through File labeling and summary sheets cost <$100 but addressed 42% of pre-intervention workflow inefficiencies. This resonates with Ramly et al.’s [14] argument for “pragmatic, low-cost implementation strategies” in resource-limited settings.  Our methodology was based on contextual factors (e.g. staff engagement) rather than purely focusing on technology, making it replicable in similar Sub-Saharan African hospitals.

Implementation Science Framework Application

Using the Implememation science approach in piloting this clinic was a unique approach, that allowed complementing a clinical intervention (the specialized clinic) with implementation strategies that helped overcome local context barriers and took advantage of existing context facilitators of implementation outcomes such as reach, and adoption as emphasized by Ramly et al [14]. The specialized prostate cancer clinic addresses a critical need locally, where the indirect social and economic costs of prostate cancer treatment usually fall back on the patient [6, 10]. From the results, the three outcomes that were tracked : clinic flow process, accuracy of diagnoses records and time to final diagnosis reflect the three main domains of the CFIR outcomes domains; Adoption, implementation, sustainment respectively [13]; these three domains have revealed a significant positive impact of the new prostate cancer service. The “sustainment” of the service domain will require further follow up and more robust documentation, that can form the basis for further implementation cycles using a similar model.

This study exemplifies how implementation science can be applied to transform clinical practice. By focusing on feasible, context-specific adjustments—staff engagement, file labeling, and clinic scheduling—we achieved measurable improvements in PCa care delivery at our centre. The intervention’s alignment with Zambia’s national cancer strategy further exemplifies its policy relevance, while its low-cost design offers a template for similar settings. As PCa burdens rise across sub-Saharan Africa, approaches such as these will be critical to in addressing cancer care access disparities in our region.

Limitations

Despite the positive findings, our study had several limitations.  Data reliability was compromised by inconsistent records, due to paper-based documentation; for example, less than 50% pre- intervention  time-to-diagnosis data were retrievable. Further, the CFIR cycle evaluation was not complete,  as long-term outcomes such as survival, cost-effectiveness, and sustainment metrics were not assessed and would require follow-up studies. Finally, the scalability of the model is uncertain, as its success in a single tertiary center may not translate to district or rural clinics with fewer specialists and limited resources. To address these gaps, we recommendation the formulation of local prostate cancer treatment guidelines to standardize care in specialized units, incorporating patient perspectives through patient reported outcome surveys, and rolling out the specialized clinic model to other sites with similar settings before broader implementation.