Economic burden of liver cancer: a multicenter study in Ghana

¹Department of Health Services Management and Informatics, Klintaps University College of Health and Allied Sciences, Accra, Ghana, ²37 Military Hospital, Pediatric Department, Accra, Ghana, ³Department of Health Policy Planning and Management, University of Ghana, School of Public Health, Accra, Ghana

^&Corresponding author
Ebenezer Owiredu Nkansah, Department of Health Services Management and Informatics, Klintaps University College of Health and Allied Sciences, Accra, Ghana

Introduction    

Liver cancer remains a major global health problem, ranking sixth in incidence and second among causes of cancer-related death in men worldwide, with an estimated 866,136 new cases and 758,725 deaths reported annually [1]. The burden is disproportionately high in sub-Saharan Africa, which accounts for a substantial share of global cases, driven largely by late diagnosis, limited healthcare access, and a high prevalence of chronic hepatitis B virus infection [2,3]. In Ghana, liver cancer has become the leading cause of cancer-related mortality, with approximately 3,452 new cases and 3,166 deaths reported annually [3]; affected patients tend to be diagnosed relatively young and at advanced disease stages [4], a pattern that often coincides with the most economically productive years of life [5].

Beyond its clinical toll, liver cancer imposes a considerable economic burden on patients, households, and health systems through both direct and indirect costs [6,7]. Direct costs comprise medical expenditure, such as consultations, diagnostic investigations, imaging, and medicines, and non-medical expenditure, such as transport and subsistence; indirect costs capture the value of productivity lost to illness, typically estimated through the human capital approach [8,9]. In other settings, hospitalization costs of up to US$7,863 per patient have been documented, with surgical interventions accounting for a substantial share of inpatient expenditure [10,11]; in high-income countries such as the United States, median monthly direct medical costs have ranged from US$21,282 to US$35,395 [12], while diagnostic imaging and laboratory testing alone have been reported to account for 20-25% of initial-phase costs [13]. Indirect costs are particularly pronounced in low- and middle-income countries, where productivity losses from premature mortality can translate into hundreds of millions of dollars in foregone output [14], reflecting the tendency of the disease to strike patients during their peak earning years [5].

Direct costs of liver cancer care in Ghana, including hospitalization, investigations, and medication, are largely borne out-of-pocket because of limited health insurance coverage for cancer care, while indirect costs from lost productivity affect a predominantly economically active patient population [2,15]. Despite this, liver cancer in Ghana remains poorly characterized from a cost-of-illness perspective; most local research has focused on the clinical and epidemiological profile of the disease [16], with little empirical evidence on the magnitude and composition of the financial burden borne by patients and their families. This evidence gap limits the ability of policymakers and health system planners to design appropriately targeted financial-protection mechanisms, insurance benefit packages, and resource-allocation strategies for liver cancer care in the country [17].

This study therefore estimated the direct and indirect costs of managing liver cancer among patients attending oncology departments in three teaching hospitals in Ghana, at the perspective of the patient/household, with the aim of generating empirical evidence to inform health-financing policy and clinical decision-making for liver cancer care.

Methods     

Study design and setting: a quantitative, cross-sectional, prevalence-based cost-of-illness design was conducted from the patient/household perspective to estimate the direct and indirect costs of managing liver cancer [8,18]. This design provides a cross-sectional view of all costs related to managing a condition for affected persons at a single point in time and is particularly useful for estimating the cost of less prevalent conditions [19,20]. The study was conducted in the oncology/gastroenterology clinics of the Korle-Bu Teaching Hospital (KBTH), Cape Coast Teaching Hospital (CCTH), and Tamale Teaching Hospital (TTH), the three institutions that serve as the principal referral centers for cancer care in southern, coastal, and northern Ghana, respectively.

Study population and sampling: the target population comprised adult (18 years and older) patients receiving liver cancer treatment at the oncology/cancer units of the three hospitals. Patients of either sex who had received treatment at the facility for at least one month at the time of data collection were eligible, a criterion adopted to ensure that relevant costs were well captured given the typically short survival of patients with this condition. Patients who declined consent, were too ill to be interviewed, had communication or cognitive difficulties, or were below 18 years were excluded, as were those who had taken part in the instrument pre-test. The three hospitals were purposively selected for their dedicated oncology services and role as national/zonal referral centers [21,22], while a census approach was used to recruit all eligible patients who attended the three hospitals' gastroenterology/oncology clinics over four months (April-July 2025) [23-25]. A total of 121 patients were interviewed: 34 from CCTH, 51 from KBTH, and 36 from TTH. During the four-month study period (April-July 2025), all patients presenting with liver cancer at the three oncology clinics were screened using a census sampling approach. Of the 152 patients screened, 21 were excluded: seven were too ill or incapacitated to complete the interview, two were younger than 18 years, five had less than 30 days of treatment history at the facility, and seven had participated in the pre-testing of the study instrument. Of the 131 eligible patients approached, 10 declined to participate, yielding a final sample of 121 patients and a participation rate of 92.4%.

Data collection: data were collected through structured, interviewer-administered questionnaires covering sociodemographic characteristics, health/treatment status, direct cost, and indirect cost, programmed onto the KoboToolBox electronic data-capture platform. Interviews were conducted face-to-face in English, Akan, or Ga by three trained research assistants under the supervision of the principal investigator, following a one-day training session and a seven-participant pre-test conducted at the Komfo Anokye Teaching Hospital. Written or thumb-printed informed consent was obtained from all participants before enrolment.

Estimation of direct cost: direct cost was estimated by summing all direct medical and non-medical costs of managing liver cancer over the preceding month. Direct medical cost comprised consultation, medicines, laboratory investigations, imaging, and admission costs reported by patients for their most recent facility visit; each cost component was multiplied by the patient's reported average number of clinic visits per month to derive a monthly estimate. Direct non-medical cost comprised transportation, food and drink, and other incidental costs incurred by patients and caregivers in seeking care. Total direct cost was obtained by summing all medical and non-medical cost components.

Estimation of indirect cost: indirect cost was estimated using the human capital approach, which assigns a monetary value to lost economic productivity attributable to illness [8,26]. Productivity loss was calculated by summing waiting time at the hospital, travel time to and from the hospital, and absenteeism from work for both patients and caregivers over one month, and valuing this lost time at the 2025 Ghana national daily minimum wage of GHS19.97 [27]. Productivity losses associated with illness, healthcare-related travel, and caregiving were valued using the 2025 Ghana National Daily Minimum Wage of GHS 19.97 per day. This standardized wage rate was applied to estimate the opportunity cost of lost productive time among both employed and non-employed participants. Among the 77 employed participants (63.6%), the wage represented a conservative estimate of lost earning capacity. For the 44 non-employed participants (36.4%)-comprising those who were unemployed, retired, or students-the same wage rate was applied to assign an economic value to time lost from unpaid domestic activities, education, caregiving, and other productive activities, rather than assigning a zero value to non-market time. Similarly, time contributed by informal caregivers was valued at GHS 19.97 per day to reflect the opportunity cost of foregone household production and other productive activities.

All cost estimates were standardized to a one-month horizon; a shorter horizon was selected because the poor prognosis and short survival typical of liver cancer in Ghana [28] make longer costing periods less representative of the full patient cohort. After all, a shorter recall period reduces recall bias in self-reported expenditure and permits valid comparison across patients with differing clinic-visit frequencies. Economic burden was defined as the sum of direct and indirect costs. Costs in Ghana cedis (GHS) were converted to United States dollars (US$) using the Bank of Ghana interbank exchange rate of US$1 = GHS10.28, as of June 2025 [29].

Sensitivity analysis: a multiple-way sensitivity analysis was conducted by simultaneously varying the costs of imaging, medicines, and the national minimum wage, the three parameters identified as the major drivers of direct medical and indirect cost, by 25% in each direction, to assess the robustness of the cost estimates against plausible fluctuations in hospital pricing, drug costs, and wage policy.

Data processing and analysis: data were captured in Microsoft Excel, cleaned, coded, and exported to Stata version 17 (StataCorp, College Station, TX) for analysis. Descriptive statistics (frequencies, percentages, means, standard deviations, medians, and interquartile ranges) were used to summarize sociodemographic characteristics and cost outcomes, disaggregated by facility and cost category, and presented with 95% confidence intervals where applicable.

Ethical considerations: ethical approval was obtained from the Institutional Review Boards of the Cape Coast Teaching Hospital (CCTHERC/EC/2025/052), the Korle-Bu Teaching Hospital (KBTH-STC 00016/2025), and the research and development directorate of the Tamale Teaching Hospital (TTH/R&D/SR/25/133). Permission was also obtained from the heads of the respective oncology departments. Participation was voluntary, confidentiality was maintained through questionnaire coding, no compensation was provided, and participants faced no risk to their ongoing care for declining or withdrawing from the study.

Results     

Sociodemographic and clinical characteristics: a total of 121 patients with liver cancer were recruited from the three teaching hospitals: Cape Coast Teaching Hospital (CCTH) (28.1%, n=34), Korle-Bu Teaching Hospital (KBTH) (42.1%, n=51), and Tamale Teaching Hospital (TTH) (29.8%, n=36). Overall, 56.2% of participants were male, and 69.4% were married. The mean age was 48.8 years (SD=12.2; 95% CI: 45.70-50.37), with 59.5% aged 40-59 years. At TTH, 30.6% (11/36) of participants were aged 25-39 years. Regarding employment status, 63.6% (77/121) were employed, 21.5% (26/121) were unemployed, 9.1% (11/121) were retired, and 5.8% (7/121) were students. Among the 77 employed participants, 75 reported monthly income and were included in the income analysis; the median monthly income among these participants was GHS 3,000 (IQR=GHS 3,300), with 22.7% (17/75) earning less than GHS 1,000 per month. The remaining 44 participants, comprising those who were unemployed, retired, or students, reported no direct earned income. Educational attainment varied, with 19.0% having no formal education, 31.4% having basic education, 19.8% secondary education, and 29.8% having tertiary education. Most participants (92.6%) had health insurance coverage. Clinically, 70.2% of patients were diagnosed at stage I, 24.8% at stage II, and 5.0% at stage III (Table 1).

Direct cost of managing liver cancer: the total direct cost of managing liver cancer for one month was estimated at GHS642,770.00 (US$62,526.26). Medical costs accounted for 88.3% (GHS567,881.00; US$55,241.34) of this total, while non-medical costs accounted for 11.7% (GHS74,889; US$7,284.92). Consultation and admission costs contributed relatively less to direct medical costs across the facilities. However, admission costs represented 11.4% of direct medical costs at TTH, with a mean cost of GHS 742.92 (US$72.27) per patient, corresponding to a total cost of GHS 26,745.00. Across all study participants, the pooled mean admission cost was GHS 688.33 (US$66.95; 95% CI: 345.86-1,030.81). Among medical cost components, imaging was the leading driver, constituting 34.7% of medical costs at both CCTH and KBTH and 33.7% at TTH; patients at KBTH spent GHS84,789 (US$8,247.96) on imaging, compared with GHS79,146 (US$7,699.03) at TTH and GHS30,500 (US$2,966.93) at CCTH. Laboratory investigations made up 35.8% of medical costs at KBTH (GHS87,790; US$8,539.88), the largest single share recorded across facilities. Medicines accounted for 24.5%, 19.7%, and 36.7% of medical costs at CCTH, KBTH, and TTH, respectively, and were highest in absolute terms at TTH (GHS86,245; US$8,389.59) compared with KBTH (GHS48,370; US$4,705.25) and CCTH (GHS21,555; US$2,096.79).

The mean monthly medicine cost per patient was GHS2,395.69 (US$233.04; 95% CI: 1,893.30-2,897.56), the mean imaging cost GHS2,198.50 (US$213.92; 95% CI: 1,953.87-2,443.10), and the mean laboratory cost GHS692.94 (US$67.39; 95% CI: 538.25-847.63). Mean consultation cost was GHS228.24 (US$22.20; 95% CI: 125.80-330.67), and mean admission cost was GHS688.33 (US$66.95; 95% CI: 345.86-1,030.81). Transportation was the dominant non-medical component, constituting 56.0% of non-medical costs at CCTH, 62.6% at KBTH, and 75.5% at TTH; mean monthly transport cost ranged from GHS244.72 (US$23.80) at TTH to GHS531.76 (US$51.70) at KBTH. Mean food and drink costs ranged from GHS65.14 (US$6.34) at TTH to GHS229.24 (US$22.30) at KBTH, while other incidental costs were the smallest component, with means ranging from GHS14.44 (US$1.41) to GHS88.19 (US$8.58) across facilities (Table 2).

Indirect cost of managing liver cancer: the total monthly indirect cost, estimated using the human capital approach, was GHS12,259.30 (US$1,192.54), corresponding to 613.9 productive days lost across patients and caregivers. Patients accounted for the majority of this loss, 481.9 days valued at GHS9,623.40 (US$936.30), while caregivers lost 132.0 days valued at GHS2,635.80 (US$256.40). On average, each patient lost 31.9 hours of productive time per month, valued at approximately GHS79.5; this ranged from 20.0 hours (GHS50.0) at TTH to 38.7 hours (GHS96.6) at KBTH and 34.1 hours (GHS85.2) at CCTH. Caregivers lost an average of 8.7 hours per month (GHS21.8), ranging from 6.8 hours (GHS16.9) at CCTH to 9.5 hours each at KBTH (GHS23.8) and TTH (GHS23.6). Absenteeism from work was the principal driver of productivity loss among patients, accounting for 79.7% of patient losses at CCTH, 65.6% at KBTH, and 72.1% at TTH. Among caregivers, all recorded productivity losses were attributable to travel and waiting time associated with accompanying patients to and from the hospital (Table 3).

Total economic cost of managing liver cancer: the estimated combined monthly economic burden of liver cancer across the three hospitals was GHS 655,029.30 (US$63,718.80), corresponding to a mean monthly cost of GHS 5,413.47 (US$526.66) per patient. Korle-Bu Teaching Hospital accounted for the largest share of the total burden (GHS 294,671.70; US$28,664.56), representing 45.0% of the overall cost, with a mean cost of GHS 5,777.88 (US$562.09) per patient. Tamale Teaching Hospital contributed GHS 249,127.00 (US$24,234.14), equivalent to 38.0% of the total burden, with the highest mean cost per patient at GHS 6,920.19 (US$673.09). Cape Coast Teaching Hospital accounted for GHS 111,230.60 (US$10,820.10), representing 17.0% of the total burden, with a mean cost of GHS 3,271.49 (US$318.24) per patient. Direct costs dominated the overall economic burden, contributing 98.1% of the total (GHS642,770.00; US$62,526.26; mean GHS5,312.15 per patient), while indirect costs contributed only 1.9% (GHS12,259.30; US$1,192.54; mean GHS101.32 per patient). This pattern held across all three facilities: at KBTH, direct costs constituted 99.1% of the facility total; at TTH, 97.3%; and at CCTH, 93.7%, with indirect costs forming a correspondingly larger share (6.3%) at CCTH than at the other two hospitals (Table 4).

Sensitivity analysis: base-case and multi-way sensitivity analysis results are presented in Table 5. Under the base-case scenario, the estimated monthly economic burden of liver cancer across the three hospitals was GHS 655,029.30 (US$63,718.80). Simultaneously increasing the costs of imaging and medicines and the national minimum wage by 25% increased the total burden to GHS 745,745.38 (US$72,543.32), representing a 13.8% increase. Conversely, a simultaneous 25% reduction in these parameters decreased the total burden to GHS 564,313.22 (US$54,894.28), corresponding to a 13.8% reduction. At the facility level, the 25% increase resulted in cost increases of 12.5% at CCTH (from GHS 111,230.60 to GHS 125,112.00), 11.8% at KBTH (from GHS 294,671.70 to GHS 329,496.12), and 16.9% at TTH (from GHS 249,127.00 to GHS 291,137.25). In contrast, increasing the national minimum wage alone by 25% had a negligible effect on the overall burden, increasing total costs from GHS 655,029.30 to GHS 658,094.13 (0.5%). This limited sensitivity reflects the relatively small contribution of indirect costs to the total economic burden (Table 5).

Discussion     

This study estimated the direct and indirect costs of managing liver cancer among 121 patients attending three teaching hospitals in Ghana, finding a substantial monthly economic burden dominated by direct medical expenditure.

Direct cost of managing liver cancer: the total direct cost of GHS642,770.00 (US$62,526.26) per month, with medical costs (88.3%) far outweighing non-medical costs (11.7%), confirms earlier findings that hospital-based expenditure, particularly diagnostics and medicines, constitutes the bulk of cancer-care costs [30,31]. Diagnostic services, imaging and laboratory tests combined, accounted for the largest share of medical costs in this study, consistent with evidence that diagnostics contribute 20-25% of initial cancer-care costs worldwide [32] and with reports that diagnostic out-of-pocket spending is especially burdensome in low- and middle-income countries (LMICs) [14]. This pattern likely reflects both the need for repeated investigations to confirm diagnosis and stage disease, and the tendency for patients to present late with advanced disease [5,20].

Medicine costs represented a smaller share of direct medical expenditure than has been reported in high-income settings, where therapies such as sorafenib and immunotherapies can account for over half of direct costs [12,14]. The comparatively modest medicine costs observed here may reflect limited use of cancer-specific drugs, with many patients still in the investigative phase or being treated for underlying hepatitis B infection, the leading cause of liver cancer in Ghana [28,33]; however, as this study did not directly assess access to, or availability of, advanced therapies, this explanation should be regarded as a plausible interpretation rather than a confirmed finding. It is broadly consistent with evidence that lower pharmaceutical costs in LMICs can reflect restricted access to advanced therapies rather than genuine affordability [17]. Non-medical costs, though a smaller proportion of the overall direct burden, remained substantial, with transportation as the leading component, mirroring evidence that transport and subsistence costs are a critical, often underappreciated, component of cancer-care costs in LMICs [15].

Indirect cost of managing liver cancer: the indirect cost of GHS12,259.30 (US$1,192.54) per month, representing a small proportion of the overall economic burden, nonetheless reflects a real and additional financial strain on patients and caregivers already managing substantial out-of-pocket medical expenditure. Patients bore the larger share of productivity loss, driven mainly by absenteeism, while caregiver losses were entirely attributable to travel and waiting time, a pattern consistent with evidence that travel to centralized oncology centers and time away from work are major drivers of indirect cancer costs in LMICs [34-37].

The relatively small proportional contribution of indirect costs observed in this study contrasts with reports from some LMIC settings where productivity losses constitute a much larger share of the total cancer burden [14]. This divergence may be explained by the one-month costing horizon adopted here, which is shorter than the six- to twelve-month horizons used in several comparator studies and would be expected to understate the cumulative, longer-term productivity losses associated with the disease [14]. The relatively limited use of advanced systemic therapies in this setting, which shortens the duration of treatment-related absenteeism compared with high-income settings where prolonged regimens are more common, may also contribute to this pattern [5].

Total economic cost of managing liver cancer: the comparatively low contribution of medicine costs to the overall burden, in contrast to high-income settings where systemic therapies dominate expenditure, may suggest constrained access to advanced cancer therapeutics rather than genuine affordability, although this study did not directly examine medicine access or availability [38]. The small relative share of indirect costs observed here, by contrast, diverges from some LMIC evidence suggesting that productivity losses can constitute 15-30% of the total cancer burden [17]. As noted above, the monthly costing horizon used in this study, together with comparatively low waiting-time costs, likely accounts for this divergence; structured appointment systems in Ghanaian oncology clinics may help mitigate waiting-time losses, although travel-related losses remained substantial, particularly for caregivers. Together, these findings point to the need for policy reforms that expand diagnostic and medicine coverage under national health insurance, and that decentralize oncology services to reduce travel-related costs, to ease the dual direct and indirect financial burden borne by households.

Theoretical application: human capital theory: the findings of this study are consistent with Human Capital Theory as a framework for interpreting the economic burden of liver cancer. The high out-of-pocket expenditure observed on diagnostics and treatment illustrates the erosion of household financial resources through healthcare consumption, consistent with evidence that cancer treatment costs often exceed household incomes in LMICs [8]. The productivity losses recorded among patients and caregivers, concentrated among patients in economically active age groups, support the theory's assertion that health represents a form of capital whose deterioration directly undermines economic output [39,40], and align with the broader argument that health capital underpins household stability and national productivity [41].

Limitations: several limitations should be considered when interpreting these findings. The relatively small sample, drawn from three tertiary hospitals, may not fully represent the broader population of liver cancer patients in Ghana, and cost estimates should be interpreted with caution when applied beyond the study sites. Much of the cost information was self-reported by patients and caregivers, raising the possibility of recall bias, although this was mitigated by triangulation with hospital records where available. The one-month costing horizon, while improving comparability across facilities and patients with differing visit frequencies, likely underestimates the cumulative, longer-term economic burden of liver cancer, particularly for patients undergoing extended treatment or recurrent hospitalization [42]. Finally, the limited body of published cost-of-illness research specific to liver cancer in sub-Saharan Africa constrained opportunities for deeper regional comparison, underscoring the need for more comprehensive cost analyses in the region.

Despite these limitations, this study makes an important contribution as one of the few to quantify both the direct and indirect economic burden of liver cancer in Ghana and, to the authors' knowledge, the first to provide empirical monthly household cost estimates across multiple tertiary facilities in the country.

Conclusion     

Liver cancer imposes a substantial monthly economic burden on patients and households in Ghana, driven overwhelmingly by direct, largely out-of-pocket, medical costs for diagnostics, medicines, and hospitalisation, with a smaller but still meaningful contribution from productivity losses among patients and caregivers. Most patients in this study were within the economically active age group, meaning that even the comparatively smaller indirect cost component represents a real loss of household income on top of substantial direct expenditure. The concentration of the burden in diagnostic and medicine costs, combined with the absence of comprehensive financial protection for cancer care under the National Health Insurance Scheme, deepens the financial strain faced by patients and their families.

Recommendations

Based on these findings, the following are recommended: 1) The National Health Insurance Scheme should expand coverage for liver cancer care to include diagnostic investigations, essential medicines, and treatment services, to protect patients and households from catastrophic out-of-pocket expenditure. 2) The Ministry of Health and Ghana Health Service should strengthen liver cancer prevention and early detection through intensified hepatitis B vaccination and expanded screening of high-risk groups, which would reduce the diagnostic and treatment costs associated with late-stage presentation. 3) The government should decentralize oncology and diagnostic services to regional and selected district hospitals to reduce the non-medical, travel-related costs borne by patients and caregivers and to bring care closer to where patients live. 4) Health facilities should adopt structured appointment systems and explore telemedicine services to reduce waiting times, travel costs, and the associated indirect economic losses to patients and caregivers. 5) Future studies should examine the cost-effectiveness of various liver cancer treatment approaches in Ghana and assess the longer-term direct and indirect economic burden of the disease using extended costing horizons.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Ebenezer Owiredu Nkansah conceived and designed the study, collected and analysed the data, and drafted the manuscript. Rosina Enyonam Daitey collected and analysed the data and reviewed the manuscript. Richmond Owusu supervised the project, analysed the data, and edited the manuscript. All the authors have read and approved the final version of this manuscript.

Acknowledgments     

The authors thank the management and staff of the oncology/gastroenterology departments of the Korle-Bu, Cape Coast, and Tamale Teaching Hospitals, and the patients who participated in this study.

Tables     

Table 1: sociodemographic and clinical characteristics of study participants (N = 121)

Table 2: direct cost of managing liver cancer by facility, one-month recall period

Table 3: indirect cost of managing liver cancer (productivity loss, human capital approach), one-month recall period

Table 4: total economic cost (direct and indirect) of managing liver cancer by facility

Table 5: multi-way sensitivity analysis of the total economic cost of managing liver cancer (±25% variation in imaging, medicine, and minimum-wage costs)

References