Acceptability, Feasibility, Appropriateness, and Willingness to Use a Family Practice Model Among University Staff Accessing Outpatient Services at a University-Owned Hospital in Ghana

Abstract

Background

Primary healthcare systems in sub-Saharan Africa face numerous challenges that require innovative solutions. Key stakeholder perspectives are essential in the successful implementation of any interventions. This study assessed the acceptability, feasibility, and appropriateness of a family practice model (FPM) at a university hospital in Ghana.

Methods

This was a cross-sectional study conducted among 345 university staff who had accessed outpatient services at the University Hospital, Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana, before the implementation of the FPM. A standardised pre-tested questionnaire was developed as a Google Form and distributed to university staff via all communication channels. Factors independently associated with willingness to accept the FPM were assessed using multivariable logistic regression analysis.

Results

The mean age of the study participants was 45.5 (±10.5) years. Approximately 92.8% of study participants perceived the FPM implementation as acceptable and feasible, and 87.8% rated it as appropriate. About 92.2% of the study participants were willing to accept the FPM. In the main multivariable analysis, making 11 or more hospital visits per year was associated with lower odds of willingness to accept the FPM (AOR: 0.19, 95%CI: 0.05 – 0.76), while reporting frustration with the consultation process was associated with higher odds of willingness to accept the FPM (AOR: 8.70, 95%CI: 1.96–38.69). These associations remained in a sensitivity analysis.

Conclusion

The FPM was perceived as acceptable, feasible and appropriate and willingness to accept it was high among the study participants. These findings reflect perceived attitudes and should not be interpreted as evidence of implementation readiness or actual uptake.

Introduction

The United Nations Sustainable Development Goals (SDGs) emphasise achieving Universal Health Coverage (UHC), with access to essential health services as a bedrock. ¹ UHC requires access to a doctor and high-quality healthcare services when needed. Primary Health Care (PHC) has been extensively studied as a key approach to achieving UHC.^2,3 According to the Alma-Ata Declaration, PHC is essential health care based on practical, scientifically sound, and socially acceptable methods and technologies, made universally accessible to individuals and families in the community through their full participation, at a cost the community and country can afford. ⁴

Ghana’s healthcare system is made up of three levels, including primary, secondary, and tertiary care, with PHC being the cornerstone of its operation. ⁵ PHC is organised at three levels: district (the highest), sub-district, and community, and is delivered through Community-based Health Planning and Services compounds, health centres, and district hospitals.^5-7 These services encompass preventive, promotive, and basic curative care, serving as the initial point of contact for many Ghanaians, particularly in rural and low-resource settings,^5-7 enhancing access to UHC.

Achieving UHC requires leadership, innovation, effective stakeholder involvement, commitment to change, and a well-designed, efficient healthcare system centred on the family. A family is defined as a group of people who are related by birth, by law, or by choice. ⁸ Understanding family dynamics is crucial for appreciating the natural history and epidemiology of illnesses that affect health outcomes. ⁹ The fields of family medicine and general practice are characterised by serving as the patient’s initial point of contact with the healthcare system, providing enhanced access to practitioners, and treating all medical conditions regardless of the patient’s age, sex, or other characteristics. Family practice, as a clinical discipline, has evolved since its inception in Canada and Great Britain in the 1960s, culminating in its recognition as a speciality in the United States in 1969. ¹⁰ It was popularised in Europe and other regions before its introduction in Africa. ¹¹

In Ghana, the increasing adult population has been accompanied by a surge in medical conditions, amid limited healthcare resources. ¹² Outpatient services are also burdened by challenges such as long waiting times, a lack of personalised care, low patient satisfaction, poor continuity of care, and an inefficient appointment system. ¹³ There is a growing awareness among the public and policymakers of the need to strengthen the healthcare system to promote healthier individuals, families, and communities at lower cost and improved outcomes. ¹⁴ The World Academy of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians (WONCA) recommends that every family should have access to a family doctor. ¹⁵ However, this is considered lofty and beyond reach in most sub-Saharan countries, including Ghana.

In response to these challenges, the management of the University Hospital, Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana, initiated a pilot Family Practice Model (FPM) in 2024 for university staff and their dependents, who are a major stakeholder group at the facility. The FPM was designed to enhance continuity of care, improve patient-provider relationships, and facilitate access to healthcare services within the outpatient department during non-emergency cases. Unlike routine outpatient care, in which patients may see different doctors at each visit, the FPM focuses on a structured, family-level provider relationship. A university staff member registers themselves and their family members and selects up to three designated doctors, with one serving as the main doctor. Individuals primarily access non-emergency care from their main doctor and can also access care from either of the two designated doctors when the main one is unavailable. All the doctors’ schedules are made available on an online booking platform at the beginning of every month, allowing individuals to select available appointments. The FPM, is therefore distinct from routine outpatient care, where patients do not have a preference for a specific doctor may depend mainly on provider availability. This is because the model establishes a structured family-level relationship with three designated providers, centred on a main doctor, with access to two alternative clinicians. The FPM therefore seeks to promote continuity and familiarity with a provider, as well as flexible access to alternative designated providers. The choice of the university staff and dependents was informed by the need to pilot this model in a small, manageable population before potential expansion to a broader population.

Based on feedback from patients, particularly university staff, at the University Hospital, KNUST, including reports of long wait times, service delays, and a limited continuity of care, the FPM offers a novel approach to address these gaps. Before implementation, it is essential to have a nuanced understanding of the potential beneficiaries’ perceived acceptability, feasibility, and appropriateness, as these may be key determinants of the model’s uptake and sustainability. Hence, this study assessed the acceptability, feasibility, and appropriateness of the proposed FPM among university staff at the University Hospital, KNUST, and examined factors associated with willingness to use the model.

Although FPM has been acknowledged to be very useful in enhancing health outcomes through continuity and comprehensiveness of care in other jurisdictions,^16,17 it is not yet integrated into Ghana’s healthcare system as a formal approach to PHC delivery. Ghana’s quest to achieve UHC may benefit from bolstering PHC through a patient-centred approach such as the FPM. However, data on how such models are perceived in Ghana is scarce. To the best of the authors’ knowledge, this is among the first studies to provide empirical evidence on the acceptability, feasibility, and appropriateness of the FPM in Ghana, offering data to inform potential scale-up and also guide policy development aimed at strengthening PHC services in the local context.

Methods

Study design and setting

This study employed a cross-sectional design to assess the acceptability, feasibility, appropriateness, and willingness to use the FPM. The design was appropriate because it enabled evaluation of these implementation outcomes and identification of factors associated with willingness to use the FPM among university staff. Cross-sectional designs are particularly useful for pilot assessments conducted prior to large-scale implementation. ¹⁸ The study was conducted at the University Hospital, KNUST. It is a 135-bed district-level primary care facility established in 1952, with the primary mandate to provide healthcare to university staff and their dependents, as well as students. Over the years, it has expanded its services beyond the university campus to involve a catchment population of over 350,000 within the Oforikrom Municipality, Kumasi. The hospital’s services include family medicine, internal medicine, obstetrics and gynaecology, general surgery, and other surgical specialities. The total outpatient attendance in 2022 and 2023 was 212,856 and 209,812, respectively. The hospital has 460 administrative and medical staff, comprising 42 medical officers, 177 nurses, and 241 auxiliary staff.

Description of the Family Practice Model

In 2024, the hospital management approved the implementation of the FPM with the slogan ‘A family doctor for every family’. The FPM is a structured, family-level provider enrolment intervention designed to provide continuity and convenient access to non-emergency outpatient services. A university staff member registers themselves and their family members and chooses up to three designated doctors, with one serving as the main physician under this model. While the main doctor serves as the family’s primary point of contact for non-emergency consultations, the other two designated doctors provide alternative access in the main clinician’s absence. At the beginning of every month, the doctors’ schedules are displayed on an online booking platform, allowing patients to schedule convenient available appointments. The FPM, therefore, seeks to ensure continuity, familiarity and flexibility in accessing alternative designated physicians. Overall, this seeks to facilitate more personalised and continuous care by allowing the designated physicians to become familiar with the patient’s and family’s healthcare needs and medical history. This model does not involve emergency services, and it is made known to the staff member at the time of registration.

Study population

The study participants were university staff aged 18 years and above who had accessed outpatient department services at the hospital prior to the implementation of the FPM. The University staff were chosen as the study population to pilot the FPM implementation before its potential scale-up to the broader patient population accessing care at the University Hospital.

Sample size estimation

The Charan and Biswas formula was used to estimate the sample size for this study. ¹⁹

$$\text{Sample}\hspace{0.17em}\text{size}=\frac{\hspace{0.17em}{\mathrm{Z}}^2\left(\mathrm{P}\right)\hspace{0.17em}\left(1-\mathrm{P}\right)}{{\mathrm{E}}^2}$$

Where Z is the 95% confidence interval with a corresponding Z score of 1.96, E is the error margin, and P is the proportion of the population satisfied with the quality of service. Therefore, using a 95% confidence interval, a 5.0% margin of error, and a 70% satisfaction with the quality of care received in healthcare institutions in Ghana, ²⁰ an estimated sample size of 323 was obtained. With a 10% non-response rate, a total of 345 participants were recruited for the study.

Sampling and data collection

Study participants were recruited using a voluntary response sampling approach. An electronic questionnaire was created using Google Forms and distributed via all staff communication channels, including WhatsApp, Telegram, and official university emails. Each participant received a secure, one-time-use link to prevent multiple submissions. Additionally, participants were asked to indicate whether they were university staff or dependents before being given access to the survey. While these measures were employed to reduce the likelihood of ineligible participants, we acknowledge that online surveys risk challenges, such as unintended participants. ²¹ Responses were received until the estimated sample size was achieved. The voluntary response method was chosen for this study to facilitate recruitment of the defined pilot population through existing university staff communication platforms. The study was conceptualised to derive preliminary evidence on perceptions of the FPM among university staff who had previously accessed care at the hospital, to guide the potential future implementation and scale-up. Given the non-probability sampling approach used for participant recruitment, the sample was not designed to be representative of all patients attending the University Hospital.

The questionnaire (see Supplementary File 1) was developed de novo by the authors, informed by a review of relevant literature and expert input. Content validity was improved through consultations with a Family Physician Consultant, two Family Medicine residents, an Epidemiologist, and a Biostatistician who reviewed the questionnaire for accuracy, clarity, and lucidity. The questionnaire was pretested among fifteen patients accessing outpatient services at another facility, and necessary adjustments were made before data collection.

The questionnaire was used to collect data on participants’ sociodemographic characteristics, including age, marital status, gender, and educational level. Data on participants’ experiences at the hospital, such as the number of visits per year and whether they have a specific doctor who attends to them, were also collected. Data on the perceived challenges and benefits of the FPM were also measured.

Acceptability, appropriateness, and feasibility of the FPM were each measured using single-item binary (Yes/No) questions capturing participants’ perceptions of its implementation, while willingness to use/accept the FPM was measured similarly. This approach was employed given the exploratory and pilot nature of the study, where brief, simple, and easily explainable measures are useful for quickly and actionably collecting insights into new care models.^22,23 Although single-item measures restrict full psychometric assessments, they reduce the burden on respondents, improve response rate, and are considered particularly useful for early-stage implementation research. ²³ These items were adopted to provide an exploratory assessment of participants’ perceptions rather than a psychometrically validated evaluation of the three implementation outcomes.

Before the questions, we provided a short introductory paragraph that explained, in simple, clear terms, the definitions of acceptability, feasibility, and appropriateness for the FPM. Your thoughts on the proposed FPM are asked in the following questions. Please indicate whether you perceive the FPM as acceptable (satisfactory or agreeable), feasible (can be implemented successfully), and appropriate (suitable for the University Hospital). The following questions were asked to measure the acceptability, feasibility, and appropriateness of the FPM: “Do you consider the FPM acceptable?” “Do you consider the FPM feasible?”, and “Do you consider the FPM appropriate in the University Hospital?” Regarding the willingness to accept the FPM, it was measured as “Under the FPM, each person or family would regularly see one main doctor for check-ups, follow-ups, and minor illnesses. Would you be willing to receive care under this model?”

Definitions of terms

Feasibility in this study was operationalised as the perceived ability of the FPM to be effective or successfully implemented at the University Hospital. ²⁴

Acceptability in this study was operationalised as the extent to which potential beneficiaries of the FPM, if implemented at the University Hospital, perceived the intervention to be satisfactory.

Appropriateness in this study was defined as the perceived suitability or compatibility of the FPM at the University Hospital.

Willingness to accept the FPM in this study was operationalised as participants’ readiness or intention to use or enrol in the intervention if it were implemented at the University Hospital.

Data analysis

Data were entered into Stata version 16, cleaned, and verified. Categorical variables were coded as factors, and the outcome variable, ‘willingness to accept the FPM’, was coded as ‘1’ for ‘yes’ and ‘0’ for ‘no’. Data were presented in tables, and proportions were used to describe categorical variables for participants’ characteristics, while means and standard deviations were used to summarise continuous variables. All analyses were conducted on complete cases.

Factors associated with willingness to accept the FPM were identified using binary logistic regression. First, a bivariate logistic regression analysis was conducted to identify crude associations. Secondly, a multivariable analysis was conducted, adjusting for significant (p < 0.05) and conceptually important variables in the crude analysis, to identify independent predictors of willingness to accept the FPM. A chi-square analysis was conducted to examine the relationships among feasibility, acceptability and appropriateness. A conceptual overlap was observed between acceptability and appropriateness, while feasibility showed moderate distinction. Consequently, Feasibility and appropriateness were examined only in crude analyses and were not included in the multivariable models because of their conceptual proximity to willingness to accept the FPM and their potential to be modelled alongside closely related constructs. Acceptability was excluded from the main model for the same conceptual reason and was included only in a sensitivity analysis to assess the robustness of associations observed in the main model.

The Variance Inflation Factor (VIF) was used to assess multicollinearity among key explanatory variables. The results for the main model showed no evidence of multicollinearity among the explanatory variables (mean VIF = 1.19, minimum VIF = 1.04, maximum VIF = 1.061). Model fit was assessed using the Hosmer-Lemeshow test, with no evidence of poor fit (p = 0.379). Similarly, no evidence of multicollinearity was observed among the explanatory variables in the sensitivity analysis (mean VIF = 1.17, minimum VIF = 1.04, maximum VIF = 1.061). There was also no evidence of poor fit (p = 0.908) in the sensitivity model.

Results

Demographic characteristics of study participants

Of the 355 study participants who were invited to complete the questionnaire, 345 did so, yielding a response rate of 97.2%. The mean age of the study participants was 45.5 (±10.5) years, and approximately 33.9% were aged 40-49 years. Over two-thirds of the study participants were male, and the majority (85.5%) were married. The majority (95.7%) of the study participants had a tertiary education (Table 1).

Table 1.

Demographic characteristics of study participants

Variables	Total n = 345 (%)
Age group (years)
< 40	112 (32.5)
40 – 49	117 (33.9)
50+	116 (33.6)
Mean Age (±SD)	45.5 (±10.5)
Sex
Male	235 (68.1)
Female	110 (31.9)
Marital status
Single	50 (14.5)
Married	295 (85.5)
Level of education
Basic	2 (0.6)
Secondary	13 (3.8)
Tertiary	330 (95.7)

SD: Standard deviation.

Descriptive patterns of acceptability, feasibility, and appropriateness of the family practice model across participant subgroups

Among participants aged 50 years or older, acceptability appeared high (94.0%), with similar patterns observed for feasibility (94.8%) and appropriateness (87.1%). Among males, acceptability appeared high (94.0%), with feasibility (92.3%) and appropriateness (87.7%) showing similar patterns. Among study participants with a tertiary education, acceptability appeared high (92.4%), with similar patterns observed for feasibility (93.0%) and appropriateness (87.3%) (Table 2).

Table 2.

Descriptive patterns of acceptability, feasibility, and appropriateness of the family practice model across participant subgroups

Variables	Acceptability		Feasibility		Appropriateness
	Yes n =320 (92.8%)	No n = 25 (7.2%)	Yes n = 320 (92.8%)	No n = 25 (7.2%)	Yes n = 303 (87.8%)	No n = 42 (12.2%)
Age group (years)
< 40	104 (92.9)	8 (7.1)	99 (88.4)	13 (11.6)	94 (83.9)	18 (16.1)
40 – 49	107 (91.5)	10 (8.6)	111 (94.9)	6 (5.1)	108 (92.3)	9 (7.7)
50+	109 (94.0)	7 (6.0)	110 (94.8)	6 (5.2)	101(87.1)	15 (12.9)
Sex
Male	221 (94.0)	14 (6.0)	217 (92.3)	18 (7.7)	206 (87.7)	29 (12.3)
Female	99 (90.0)	11 (10.0)	103 (93.6)	7 (6.4)	97 (88.2)	13 (11.8)
Marital status
Single	45 (90.0)	5 (10.0)	46 (92.0)	4 (8.0)	44 (88.0)	6 (12.0)
Married	275 (93.2)	20 (6.8)	274 (92.9)	21 (7.1)	259 (87.8)	36 (12.2)
Level of education
Basic	2 (100.0)	0 (0.0)	2 (100.0)	0 (0.0)	2 (100.0)	0 (0.0)
Secondary	13 (100.0)	0 (0.0)	11 (84.6)	2 (15.4)	13 (100.0)	0 (0.0)
Tertiary	305 (92.4)	25 (7.6)	307 (93.0)	23 (7.0)	288 (87.3)	42 (12.7)
Staff status
Permanent	303 (92.4)	25 (7.6)	305 (93.0)	23 (7.0)	289 (88.1)	39 (11.9)
Casual/contract	17 (100.0)	0 (0.0)	15 (88.1)	2 (11.8)	14 (82.4)	3 (17.7)
Category of staff
Junior staff	10 (100.0)	0 (0.0)	8 (80.0)	2 (20.0)	10 (100.0)	0 (0.0)
Senior staff	103 (90.4)	11 (9.7)	99 (86.8)	15 (13.2)	95 (83.3)	19 (16.7)
Senior Member	204 (93.6)	14 (6.4)	210 (96.3)	8 (3.7)	196 (89.9)	22 (10.1)
Staff role
Academic	177 (90.3)	19 (9.7)	186 (94.9)	10 (5.1)	173 (88.3)	23 (11.7)
Non-academic	143 (96.0)	6 (4.0)	134 (89.9)	15 (10.1)	130 (87.3)	19 (12.8)
Number of visits in a year
1 – 4	216 (92.3)	18 (7.7)	217 (92.7)	17 (7.3)	205 (87.6)	29 (12.4)
5 – 10	72 (93.5)	5 (6.5)	69 (89.6)	8 (10.4)	66 (85.7)	11 (14.3)
11+	17 (100.0)	0 (0.0)	17 (100.0)	0 (0.0)	16 (94.1)	1 (5.9)
Frustration with the consultation process
Yes	128 (94.8)	7 (5.2)	126 (93.3)	9 (6.7)	122 (90.4)	13 (9.6)
No	192 (91.4)	18 (8.6)	194 (92.4)	16 (7.6)	181 (86.2)	29 (13.8)
Have a specific doctor who attends to me.
Yes	61 (98.4)	1 (1.6)	59 (95.2)	3 (4.8)	59 (95.2)	3 (4.8)
No	259 (91.5)	24 (8.5)	261 (92.2)	22 (7.8)	244 (86.2)	39 (13.8)

Note. Percentages are row percentages. Some subgroups have small sample sizes and should be interpreted with caution; Findings are presented descriptively, and no statistical comparisons were performed.

Willingness to accept the family practice model, perceived potential challenges, and benefits

The majority (92.2%) of the study participants were willing to accept the FPM. The study participants reported increased workload for doctors (53.3%), long waiting time (35.7%), and over-familiarity between doctors and patients (34.8%) as some of the perceived potential challenges with the FPM. Improved doctor-patient relationships (80.6%), enhanced continuity of care (80%), and improved quality of care (72.8%) were among the perceived potential benefits (Table 3).

Table 3.

Willingness to accept the family practice model, perceived potential challenges, and benefits

Variables	Frequency, N	Percentage, %
Willingness to accept FPM
Yes	318	92.2
No	27	7.8
Perceived potential challenges with the FPM
Long waiting time	123	35.7
Increased workload for doctors	184	53.3
Over-familiarity between doctor and patient	120	34.8
Increased cost of patient care	75	21.7
Frustrations in the absence of the doctor	5	1.5
Perceived potential benefits of the FPM
Short waiting time	133	38.6
Improved doctor-patient relationship	278	80.6
Enhanced continuity of care	276	80.0
Improved quality of care	251	72.8
Enhanced patient satisfaction	234	67.8
Improved patient adherence to treatment	190	55.1
Confidentiality of information about the patient	1	0.3

Factors associated with willingness to accept a family practice model

In a bivariate logistic regression analysis, the number of visits in a year, having a specific doctor, frustration with the consultation process, acceptability, and appropriateness were associated with willingness to accept the FPM (Table 4).

Table 4.

Factors associated with willingness to accept a family practice model among study participants

Variables	Crude Model		Main Model		Sensitivity Model
	Crude OR (95%CI)	p-value	Adjusted OR (95%CI)	p-value	Adjusted OR (95%CI)	p-value
Age group (years)
< 40 (ref)	1.00	​	1.00	​	1.00	​
40 – 49	0.80 (0.29 – 2.23)	0.669	0.90 (0.30 – 2.77)	0.860	1.09 (0.33 – 3.61)	0.882
50+	0.64 (0.24 – 1.70)	0.369	0.98 (0.31 – 3.10)	0.975	1.15 (0.34 – 3.92)	0.825
Sex
Male (ref)	1.00	​	1.00	​	1.00	​
Female	0.66 (0.29 – 1.47)	0.306	0.65 (0.27 – 1.54)	0.325	0.73 (0.29 – 1.84)	0.501
Marital status
Single (ref)	1.00	​	1.00	​	1.00	​
Married	1.38 (0.50 – 3.83)	0.537	1.42 (0.46 – 4.39)	0.539	1.20 (0.36 – 4.05)	0.768
Number of visits in a year
1 – 4 (ref)	1.00	​	1.00	​	1.00	​
5 – 10	0.73 (0.29 – 1.86)	0.514	0.78 (0.30 – 2.04)	0.607	0.76 (0.27 – 2.12)	0.598
11+	0.24 (0.07 – 0.82)	0.022	0.19 (0.05 – 0.76)	0.019	0.13 (0.03 – 0.56)	0.006
Have a specific doctor who attends to me
Yes (ref)	1.00	​	1.00	​	1.00	​
No	2.50 (1.07–5.86)	0.035	1.79 (0.69 – 4.68)	0.223	3.16 (1.09 – 9.19)	0.035
Frustration with the consultation process
No (ref)	1.00	​	1.00	​	1.00	​
Yes	8.99 (2.09-38.60)	0.003	8.70 (1.96 – 38.69)	0.004	8.32 (1.81 – 38.13)	0.006
Acceptability
No (ref)	1.00	​	-	-	1.00	​
Yes	7.46 (2.86-19.47)	<0.001	-	-	13.68 (4.29 – 43.59)	<0.001
Feasible
No (ref)	1.00	​	-	-	-	-
Yes	1.68 (0.47 – 6.02)	0.425	-	-	-	-
Appropriateness
No (ref)	1.00	​	-	-	-	-
Yes	4.32 (1.80-10.39)	0.001	-	-	-	-

Note. Feasibility and appropriateness were examined only in crude analyses and were not included in the multivariable models because of their conceptual proximity to willingness to accept the FPM and the potential for modelling constructs closely related to it. Acceptability was excluded from the main model for the same conceptual reason and was included only in a sensitivity analysis to assess the robustness of associations observed in the main model. Education was excluded from the multivariable model due to sparse cells and perfect prediction, resulting in non-estimable and unstable regression estimates. OR, odds ratio; CI, confidence interval. The final multivariable models included 328 observations.

In the main multivariable model, after adjusting for significant and conceptually important covariates, factors such as the number of visits per year and frustration with the consultation process remained significantly associated with willingness to accept the FPM. The study participants who reported 11 or more hospital visits in a year had approximately 81% lower odds of willingness to accept the FPM compared with those reporting 1-4 visits per year (AOR: 0.19, 95%CI: 0.05 – 0.76). In contrast, participants who reported frustration with the consultation process had about 9 times higher odds of willingness to accept the FPM compared with those who did not (AOR: 8.70, 95%CI: 1.96–38.69) (Table 4).

In a sensitivity analysis additionally adjusting for acceptability of the FPM, the findings for the number of visits per year (AOR: 0.13, 95%CI: 0.03 – 0.56) and frustration with the consultation process (AOR: 8.32, 95%CI: 1.81 – 38.13) showed similar directions and magnitudes of association. In the sensitivity model, having no specific doctor was also associated with about three times higher odds of willingness to accept the FPM (AOR: 3.16, 95%CI: 1.09 – 9.19), although this was not significant in the main model. Acceptability was associated with higher odds of willingness to accept FPM in the sensitivity model (AOR: 113.68, 95%CI: 4.29 – 43.59) (Table 4).

Discussions

This study found a high perceived level of feasibility of implementing the FPM at the University Hospital. These findings reflect participants’ perceptions of the FPM before its implementation, and not its true operational feasibility or implementation outcomes. It is worth noting that, although perceived feasibility may be important for future health system planning, our study did not examine the effect of the FPM on health outcomes, and the strengthening of PHC, or progress towards UHC. For individuals to consider an intervention feasible, they may evaluate various dimensions, including technical aspects such as the availability of human resources and logistics. Therefore, the fact that the majority of study participants considered the FPM feasible suggests they anticipated it could be implemented, though this does not necessarily mean its implementation at the University Hospital would be straightforward in practice. This finding provides an initial indication of favourable perceptions of the proposed FPM, although interpretation is limited by the use of single-item binary measures. The University Hospital currently has two certified Family Physicians and four trainees (residents), and is accredited by the Faculty of Family Medicine, Ghana College of Physicians and Surgeons, to train residents in Family Medicine. Furthermore, the hospital runs a special family medicine clinic every Monday. While these facility-related factors may plausibly support the implementation of the FPM, their direct influence on participants’ perceptions was not examined in this study.

Although the majority of study participants indicated that the FPM was feasible, a minority expressed reservations. The possibility of increased physician workload was a frequently perceived issue. Assigning specific doctors to individual patients may be perceived as increasing staffing burden and, if poorly managed, could create challenges to implementation. This suggests that, although the FPM is not viewed as excessively complex, some respondents did not consider it easy to implement within the current system at the University Hospital. These findings highlight issues raised by participants and may be key to guiding the implementation process; however, our study did not evaluate current staffing capacity or operational feasibility. Engaging stakeholders, particularly medical staff, may help address perceived concerns, including workload, in future implementation efforts.

A high level of perceived acceptability for implementing the FPM was found in this study. This indicates that most participants reported they considered the FPM acceptable; however, these findings reflect anticipated views rather than actual acceptance or behaviour following implementation. The relatively high perceived acceptability of the FPM among study participants may reflect its emphasis on establishing a structured relationship with a designated main doctor, while also providing access to two alternative doctors. This opportunity may provide a balance between continuity, familiarity with clinicians and flexibility in the absence of the main doctor. However, these reasons were not directly assessed in this study and should therefore be treated as plausible explanations rather than a demonstrated effect of the FPM. Similar findings have been reported in other countries, including South Africa, where family medicine is perceived positively by both the general public and healthcare providers. ²⁵ Consistent with this, the majority of participants in this study reported willingness to accept the FPM. Others, however, voiced doubts about implementing the FPM despite its overwhelming acceptance among most study participants. These participants preferred the former hospital system, in which patients were not assigned a specific physician for ongoing care. It is plausible that they were satisfied with the care received by various doctors in the routine outpatient system. It is worth noting that acceptability is a dynamic, context-dependent construct that may change over time.^26,27 However, changes in perceptions after implementation were not measured in this study.

This study identified a high level of perceived appropriateness for implementing the FPM in the hospital setting. Though the reasons for these findings were not assessed, participants may have considered their potential benefits before forming their views. One aspect of care highlighted in the FPM was that enrollees would be assigned a family doctor who would become familiar with their unique needs and medical history, thereby enhancing the delivery of continuous, comprehensive, and coordinated care within the hospital. This may have influenced participants’ perceptions of the appropriateness of the FPM, even though it was not specifically measured. Overall, these findings suggest that the participants perceived the FPM as a useful intervention within the hospital context, based on their anticipated perception. Similar patterns of perception of the appropriateness of family medicine have been observed in other jurisdictions, including South Africa. ²⁵ Some of the study participants were concerned about the appropriateness of implementing the FPM at the hospital. Although the reasons for these concerns were not measured directly in this study, they may be related to uncertainties about how the model will operate within the existing system. These concerns are plausible given their awareness of the hospital’s current practice, in which patients are typically seen by different physicians rather than having a designated family doctor.

In the main multivariable and sensitivity analyses, frustrations with the consultation process and number of visits per year remained significantly associated with willingness to accept the FPM, supporting the robustness of the observed associations. However, these findings should be interpreted as associations with perceived willingness among the study participants, rather than evidence of predictors of implementation or actual uptake of the FPM. These findings may suggest perceived gaps in continuity, coordination and patient-centred care within the current system, as reported by participants, rather than objectively measured challenges in service delivery. Given that the assessment of perception was conducted before implementation and did not evaluate subsequent behaviour, service use after implementation or effectiveness, the findings cannot determine whether these perceptions would translate into actual utilisation of the FPM.

Additionally, participants who reported frustrations with the current consultation were more likely to express willingness to utilise the FPM. This variable remains one of the most consistent determinants in both the main multivariable and sensitivity models. Such experiences may reflect perceived challenges in the current system, including long waiting times, poor or inadequate communication, and dissatisfaction with services, although these variables were not measured directly in this study. Hence, it is important to interpret frustration as an indicator of participants’ reported prior experience with the consultation process, rather than as a specific service-related factor or problem that was not assessed in this study. Having such perceptions may contribute to their preference for alternative care models, such as the FPM. These factors have been reported in the literature as key determinants for frustration with the consultation process.^28,29 However, the cross-sectional nature of this study limits the ability to determine whether frustrations led to a greater willingness to accept the FPM, or whether those who were more receptive to the FPM were more likely to be dissatisfied with the current consultation process.

Participants reporting 11 or more visits per year had lower odds of willingness to accept the FPM than those reporting 1–4 visits per year in both the main and sensitivity models. Although this variable remained significantly associated with willingness to accept the FPM in both models, this finding should be interpreted with caution given the relatively small number of participants in this category. Importantly, the study did not account for data on the reasons for frequent hospital visits, chronic disease burden, dependence on specialist care, continuity of care or participants’ perceptions of access to the proposed FPM. As a result, the observed association may not confidently reflect preference for the existing system or dissatisfaction with the proposed model. Alternative possible explanations, such as perceived gatekeeping and uncertainty about whether the FPM would sufficiently address the needs of frequent healthcare users, were not directly assessed. Given that these reasons were not directly measured in this study, these interpretations should be considered exploratory, requiring hypothesis generation rather than being conclusive. Future studies should therefore assess the possible influence of these explanations on willingness to accept the FPM.

In sensitivity analyses, acceptability of FPM and not having a specific doctor were associated with participants’ willingness to use FPM. It is worth pointing out that in the main analysis, not having a specific doctor was not associated with participants’ willingness to accept the FPM. However, after including acceptability in the main model as a sensitivity analysis, participants who reported not having a specific doctor had higher odds of willingness to accept the FPM. The observed change in statistical significance should be interpreted with caution and should not be considered as evidence that not having an established doctor independently increases willingness. Rather, it highlights the sensitivity of the observed association to model specification. Additionally, the observed association should not be interpreted as evidence that these participants may have experienced fragmented care or that the intervention would necessarily address such gaps, as the study did not evaluate actual continuity of care, fragmentation or satisfaction with existing providers.

The observed association between acceptability and willingness to use the proposed FPM should also be interpreted with caution, as acceptability and willingness are conceptually closely related constructs and may capture overlapping aspects of participants’ attitudes towards the model. The sensitivity analysis was thus used primarily to evaluate the robustness of the main associations, rather than to establish the acceptability, and not having a specific doctor as a factor associated with willingness. Nevertheless, this finding aligns with existing evidence that patients are more likely to use a healthcare model they perceive as improving their health outcomes.^30,31 Weiner et al also highlighted the significance of perceived acceptability as a critical factor for the adoption of an intervention. ³² Our findings in this context reflect the perceived readiness to utilise the system, not its effectiveness.

Our findings provide useful insights into participants’ perceptions of the FPM within the university hospital. The high levels of perceived acceptability, feasibility, and appropriateness of the FPM reflect participants’ views and should be interpreted as anticipated perceptions and not evidence of successful implementation. These favourable perceptions may suggest potential receptiveness to the FPM, but do not confirm that the intervention will be successful when implemented. Hence, the implication of our findings should be interpreted within this context among the study participants. Importantly, our findings may inform future research on implementing the FPM or other healthcare interventions.

Limitation

A key limitation of this study is how important implementation outcomes, such as the acceptability, appropriateness, and feasibility of the FPM, were measured. They were measured using single-item binary (Yes/No) questions. Although this approach was intentionally used to reduce respondent burden in this exploratory pre-implementation evaluation, a single binary item may not adequately capture the multidimensional nature of these constructs and does not provide the psychometric data available from validated multi-item measures such as the Acceptability of Intervention (AIM), Intervention Appropriateness Measures (IAM) and Feasibility of Intervention Measure (FIM). Therefore, the high proportions of positive responses should be interpreted with caution and may partly reflect the response format or social desirability bias. However, expert reviews and pre-testing were conducted to improve the questionnaire’s contextual significance and clarity, but they did not establish psychometric validity. Future studies should consider using validated multi-item measures, potentially adapted and culturally sensitive to the Ghanaian setting, to provide more nuanced and reliable measurement of these constructs.

The use of voluntary response sampling may be prone to potential selection bias, as individuals with prior positive experience with the healthcare system may be more likely to respond to the survey, limiting the representativeness and generalizability of the findings. Also, the use of online communication platforms for recruitment may have resulted in selection bias, as individuals with greater access to these channels or stronger opinions about the FPM may have been more likely to participate. Importantly, the study participants were restricted to university staff and were largely male and highly educated, which may limit the representativeness and transferability of the findings to other populations. Hence, the findings should be interpreted as reflecting perceptions of this cohort of university staff and should not be generalised to all patients attending the University Hospital more broadly or the general Ghanaian population. Another limitation of the study is the use of the internet to collect participants’ responses, which may have created barriers for those with limited access to or familiarity with digital technologies.

Another limitation of this study is the relatively wide confidence intervals for some adjusted odds ratio estimates, which should be interpreted with caution. It is worth noting that, as this study was a pilot and the first of its kind to examine the FPM in a hospital setting in Ghana, our objective was to generate preliminary findings using a concise and suitable tool. Future studies should include larger, more diverse populations, a probability sampling approach, and validated multi-item measures to provide more precise and generalizable findings on the implementation of the FPM.

Conclusion

Generally, the FPM was perceived by the study participants as feasible, acceptable, and appropriate within the university hospital. Nevertheless, the participants identified challenges, including increased physician workloads and the need for an efficient scheduling system. These findings highlight considerations that may require management’s attention in any future evaluation or implementation of the FPM. Our findings underscore the importance of incorporating stakeholder perspectives when designing and planning health service interventions. However, these findings should be interpreted in the context of the study’s cross-sectional design and population, given that the assessment was of perceptions and self-reported willingness among selected university staff and did not evaluate actual implementation, service utilisation, effectiveness, or scalability of the FPM. Future research should assess real-world implementation and service utilisation to determine whether the favourable perceptions and self-reported willingness observed in this study translate into actual adoption of the FPM, and further assess its applicability in diverse populations and healthcare settings.

ORCID iD

Douglas Aninng Opoku https://orcid.org/0000-0003-2321-387X

Ethical Considerations

Ethical approval for the study was obtained from the Committee for Human Research Publications and Ethics (CHRPE) at the School of Medical Sciences, Kwame Nkrumah University of Science and Technology (KNUST), Kumasi (CHRPE/AP/716/23). Before the application for ethical clearance, written approval was sought from the management of the University Hospital, KNUST. This study was conducted in accordance with the ethical principles set out in the Declaration of Helsinki. All study participants were assured of strict confidentiality during the online informed consent process. All participants who agreed to take part in the study provided consent online before being granted access to the questionnaire.

Consent to Participate

All study participants provided consent online before accessing the study questionnaire.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.*