This was a prospective cohort study that followed participants over time to assess how baseline exposures influence the development of health outcomes. The study enrolled 400 hypertensive and/or diabetic patients from four healthcare facilities in Buea, Cameroon (September 2022 – April 2024). Using interviewer-administered questionnaires, we collected data on demographic characteristics, clinical status and knowledge of CKD risk factors. Participants received standardised health education on preventive self-care practices using World Health Organization (WHO) guidelines during clinic visits.

The study was conducted at four facilities in Buea Health District, Southwest Region, Cameroon: Buea Regional Hospital, Muea Sub-Divisional Hospital, Fako Heart Hospital and Lambe Clinic (Great Soppo) (Figure 1).²⁴

Buea is situated on the eastern slopes of Mount Cameroon (4040 m), the highest peak in West and Central Africa. The region experiences distinct wet (between March and October) and dry (between November and February) seasons.

The sample size was determined using standard statistical methods for estimating a single population proportion. A 95% confidence level, a margin of error of 5% and a population proportion of 0.5 (to account for maximum variability) were used, resulting in a minimum required sample size of 385 participants. This yielded 384 participants, which we increased to 400 (accounting for 5% attrition). Facility-specific sample sizes were determined through probability proportionate to size allocation.

Participants were consecutively recruited from routine diabetes and/or hypertension clinics at the four facilities. Eligible participants were adults aged 18 years and older who had a confirmed diagnosis of hypertension (defined as systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg, or current use of antihypertensive medication) and/or type 2 diabetes (defined as fasting blood glucose ≥ 126 mg/dL or current use of glucose-lowering medication). Exclusion criteria included advanced CKD (estimated glomerular filtration rate [eGFR] < 30 mL/min/1.73 m²) and pregnancy. All participants provided written informed consent prior to enrolment.

Trained research nurses administered pre-tested, structured questionnaires during routine clinic visits. Instruments were available in both English and Pidgin (the local lingua franca) and assessed: Socio-demographic characteristics, self-care practices (medication adherence, alcohol use, physical activity), healthcare-seeking behaviours and exposure to self-care education. Questionnaires were validated through pilot testing at Limbe Regional Hospital (outside the study area).

Data were cleaned, coded and analysed descriptively (frequencies for categorical variables; measures of central tendency for continuous variables). Binary logistic regression assessed associations between five self-care outcomes (diet, exercise, weight management, medication adherence and behaviours) and three conditions (diabetes, hypertension or both), adjusting for age, sex and education. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were calculated (p < 0.05 significance). Multivariable analysis controlled for confounders. MINITAB 20.1 was used for all analyses, including model fitting and interaction testing.

Ethical clearance to conduct this study was obtained from the University of Buea (No. 2023/2080-04/UB/SG/IRB/FHS) and administrative clearance from the Southwest Regional Delegation of Public Health (No. p42/MPH/SWR/RDPH/CB.PT/681/519). The study also received facility-level approvals from all participating sites. All participants provided written informed consent prior to enrolment.

The study population comprised 400 participants with diabetes and/or hypertension from four healthcare facilities in Buea Health District, with a mean age of 60.3 years (standard deviation [s.d.] ±11.8). Females represented the majority of participants (70.8%), while males accounted for 29.2% (see Table 1). Age distribution showed that 35.3% of participants were between 61 years and 70 years old, followed by 27.3% in the 51–60 years-old age group and 15.3% aged 71 years – 80 years. Younger age groups were less represented, with only 0.8% aged 20 years – 30 years. Regarding marital status, 56.8% of participants were married, 24.0% were widowed, 17.8% were single and 1.5% were divorced or separated. Educational attainment revealed that 41.3% had primary-level education, 33.3% secondary education and 18.5% tertiary education, while 7.0% had no formal education. Occupation distribution showed 34.8% employed in the private sector, 29.5% unemployed, 21.0% farmers, 13.8% civil servants and 1.0% students. Geographically, 89.0% of participants resided in urban areas compared to 11.0% in rural communities. Financially, the majority (59.5%) earned less than 50 000 FCFA (Francs Communaute Francaise Africaine [African Financial Community {Franc}]) monthly, while only 7.0% earned above 200 000 FCFA. In terms of healthcare utilisation, over half of participants (52.3%) received treatment at Buea Regional Hospital, followed by 37.5% at Muea Sub-Divisional Hospital, with smaller proportions attending Fako Heart Centre (5.3%) and Lambe Clinic (5.0%) (see Table 1). This demographic profile highlights several important characteristics of the study population, including the predominance of middle-aged and elderly individuals, females, urban dwellers and those with limited education and income, which may influence healthcare-seeking behaviours and disease management patterns.

The study revealed a 26% prevalence of DM and a 33% prevalence of hypertension (HTN) among participants, with 41% having both conditions. Most diabetic patients (52.4%) were diagnosed ≥ 4 years prior, while 62% of hypertensive participants had HTN for ≥ 4 years. Clinical measurements showed 59% had high blood pressure and 77.75% had elevated fasting blood sugar levels (see Table 2). The data demonstrate significant disease burden, with nearly three-quarters of participants presenting abnormal glycaemic control and over half showing uncontrolled hypertension, particularly among those with longer disease duration.

Knowledge assessment revealed significant gaps in CKD awareness among participants. Only 35% recognised diabetes and hypertension as CKD risk factors, while 63% correctly identified long-term alcohol consumption as harmful. Most participants (73%) understood early detection reduces healthcare costs, but misconceptions persisted: 90% erroneously believed kidney problems could be detected through urine colour and/or smell, and 94% didn’t associate obesity with CKD risk. Notably, 98% correctly rejected water consumption as a cause, yet 72% were unaware CKD can be asymptomatic in advanced stages (see Table 3). These findings highlight critical knowledge deficits regarding both risk factors (16% recognised smoking risks) and disease presentation (only 28% knew about asymptomatic progression) (see Table 3).

The study revealed significant associations between socio-demographic factors and CKD knowledge. Older adults (41 years – 60 years) demonstrated better awareness that long-term alcohol use causes CKD (χ² = 15.67, p = 0.016), with 62.5% overall recognising this risk (see Table 4). Females showed marginally higher knowledge than males (71.3% vs. 28.7%, χ² = 399.23, p < 0.001) regarding urine-based detection misconceptions. Facility-based differences emerged, with only 1.5% at Fako Heart Centre recognising DM/HTN as CKD risks versus 16.8% at Regional Hospital (χ² = 435.26, p < 0.001). Married participants exhibited greater alcohol-CKD awareness (38.3%) than widowed individuals (11%) (see Table 4). Education significantly impacted understanding, as 94.5% with primary/secondary education believed untreated malaria causes CKD versus 5.5% with tertiary education (χ² = 407.31, p < 0.001) (see Table 4). Occupation also influenced perceptions, with 90% of private sector workers endorsing urine colour as a detection method compared to 10% of unemployed participants (χ² = 424.21, p < 0.001) (see Table 4). These findings highlight critical knowledge gaps across demographic groups regarding modifiable CKD risk factors.

The study revealed mixed self-care practices among participants (see Table 5). While dietary practices showed relatively good adherence – 97.3% avoided high-salt foods and 88.5% consumed vegetables regularly – concerning gaps existed in other areas. Only 56.5% engaged in regular exercise, with 41.8% never exercising. Medical adherence was suboptimal, with 20.5% forgetting medications and 25% missing appointments. Critical monitoring practices were particularly neglected: 84.8% lacked regular renal check-ups, and 70.2% didn’t monitor blood pressure at home. Paradoxically, 92.5% reported adding salt to food despite 97.3% claiming to avoid salty foods. Exercise frequency varied significantly, with just 10.5% exercising daily. These findings highlight a disconnect between knowledge and practice, particularly regarding cardiovascular risk monitoring and consistent lifestyle modifications. The low rates of renal monitoring (15.2%) and home BP checks (29.8%) are especially concerning given participants’ high-risk status for CKD progression (see Table 5).

The analysis revealed significant variation in self-care adherence among diabetic-hypertensive patients (see Table 6). Home glucose monitoring showed the strongest association (χ² = 8.64, p = 0.003), with 85.2% adherence among regular users versus 56.3% overall. Medication adherence approached significance (p = 0.05), with 65.9% adherence among those who occasionally forgot medications versus 56.3% among consistent users. Most practices showed no significant associations: physical activity (p = 0.86), renal check-ups (p = 0.17) and dietary salt avoidance (p = 0.17). Notably, home BP monitoring (59.9% adherent) and vegetable consumption (62.7% adherent) showed moderate but non-significant adherence patterns (see Table 6). The data suggest that while patients generally maintain basic self-care, critical monitoring behaviours like glucose checks vary significantly, potentially reflecting access barriers or knowledge gaps. These findings highlight the need for targeted interventions to improve consistent monitoring practices in this high-risk population.

The analysis revealed hypertension as a significant independent risk factor for CKD, with hypertensive patients showing 4.7 times higher crude odds (95% CI: 1.40–16.00) and 4.3 times higher adjusted odds (95% CI: 1.25–14.67) of developing CKD compared to normotensives (see Table 7). Diabetes mellitus alone showed no significant association (crude odds ratio [OR] = 0.39, 95% CI: 0.05–3.06; adjusted OR = 0.32, 95% CI: 0.04–2.74). The comorbidity of diabetes and hypertension demonstrated a non-significant trend towards increased CKD risk (adjusted OR = 1.86, 95% CI: 0.60–5.74) (see Table 7). These findings suggest that hypertension drives CKD risk in this population, while diabetes may require longer duration or poorer control to manifest significant renal effects. The persistent association after adjustment indicates hypertension’s robust relationship with CKD, independent of other measured factors.

Table 8 presents AORs, CIs and p-values for various factors associated with diabetes, hypertension and the coexistence of both conditions. Self-care management such as regular physical exercise were significantly associated with a reduced risk of hypertension (AOR: 0.07, 95% CI: 0.01 to 0.12, p = 0.015). However, this variable did not show a significant association with diabetes or with the coexistence of diabetes and hypertension, as indicated by higher p-values of 0.506 and 0.731, respectively. Other self-care behaviours, such as regular renal check-ups, salt intake, vegetable consumption and medication adherence, did not demonstrate statistically significant associations with any of the three health outcomes, as all p-values exceeded the conventional threshold of 0.05.

When investigative age groups were considered, most categories did not show significant associations with diabetes, hypertension or both. An exception was observed in the 73–83 years-old age group, where there was a statistically significant negative association with diabetes (AOR: −0.17, 95% CI: −0.32 to −0.01, p = 0.035). No significant associations were observed for this age group with hypertension or with the combined outcome (see Table 8).

For gender and education, neither female gender nor levels of education (formal, primary or secondary) were significantly associated with the risk of diabetes, hypertension or both, except for secondary education, which showed a modest but significant negative association with diabetes (AOR: −0.07, 95% CI: −0.13 to −0.01, p = 0.036).

The interaction plot (Figure 2) reveals significant trends in diabetes prevalence across age, gender and education levels. The age*gender graph reveals that the mean diabetes levels for females start at around 2.00 in the 18–29 years-old age group and gradually decrease to approximately 1.50 by the 64–94 years-old age range. In contrast, males exhibit a lower starting mean of about 1.75, which also declines to about 1.25 in older age brackets. The gender*education interaction graph indicates that individuals with tertiary education consistently show the lowest diabetes levels, with mean values near 1.50, while those with primary education have higher levels, particularly around 2.00 for younger age groups.

The interaction plot for hypertension illustrates distinct trends across age, gender and education levels (Figure 3). The interaction plot of age*gender reveals that females display a peak mean hypertension level of approximately 1.40 in the 40–50 years-old age group, while males show a more gradual increase, reaching a mean of about 1.20 in the same age range before declining slightly (Figure 3). The gender*education interaction highlights educational impacts, where individuals with tertiary education demonstrate the lowest hypertension levels, particularly among females, with a mean near 0.80, while those with primary education show higher levels, particularly in the 29–39 years-old age group, where mean hypertension is around 1.20 (Figure 3).

The interaction plot for diabetes and hypertension showcases the interplay between age, gender and education levels. The plot of age*gender reveals that females have a mean diabetes and hypertension level starting at approximately 2.00 in the 18–29 years-old age group, which decreases to around 1.60 by the 64–94 years-old age range, while males start lower at about 1.75 and decline to around 1.40 (Figure 4). The plot of gender*education illustrates that the influence of education is evident, with individuals holding tertiary education showing the lowest mean levels for both conditions, around 1.40 for females and 1.50 for males. In contrast, those with primary education exhibit higher mean levels, peaking at about 1.80 in the 40–50 years-old age group.

Our study revealed a predominantly female (70.8%), urban (89%) and older adult population (mean age 60.3 years) with limited education (41.3% primary level only) and low socioeconomic status (59.5% earning < 50 000 FCFA/month). This demographic profile aligns with patterns observed in other Cameroonian chronic disease studies.²³ The female predominance may reflect several factors: greater healthcare-seeking behaviour among women in this context,²⁵ higher prevalence of obesity and metabolic syndrome in Cameroonian women²⁶ and longer life expectancy for females in the region.²⁷

The urban concentration (89%) likely reflects healthcare access disparities, as rural populations face transportation barriers to tertiary centres, limited speciality care availability and greater reliance on traditional medicine.^28,29,30 The educational profile (only 18.5% with tertiary education) suggests health literacy challenges in medication adherence and self-care, as demonstrated in similar populations.³¹

The 41% comorbidity rate exceeds West African averages (32%),³² potentially reflecting selection bias at referral centres, true epidemiological transition in urban Cameroon and diagnostic ascertainment bias. The prolonged disease duration (≥ 4 years in 52% – 62% of participants) without adequate control (59% uncontrolled hypertension, 77.8% elevated glucose levels) suggests therapeutic inertia in primary care,³³ medication adherence challenges²⁵ and limited access to guideline-recommended therapies.³⁴

Poor recognition of diabetes and hypertension as CKD risk factors (35%) contrasts sharply with the 68% awareness in Nigerian cohorts³⁵ and 72% in South African studies.³⁶ The persistence of dangerous misconceptions (90% believed urine colour indicated kidney health) despite educational campaigns suggests current approaches fail to account for poor health literacy,³⁷ cultural health beliefs³⁸ and cognitive biases in risk perception.^39,40

The facility-based disparities (Fako Heart Centre: 1.5% vs. Regional Hospital: 16.8%) may reflect variable staff training quality,⁴¹ differences in patient education materials, or time constraints during consultations. The education gradient (tertiary-educated participants had seven-fold better knowledge) supports calls for pictogram-based education tools, community health worker interventions and mobile health solutions.^28,37,42

The salt avoidance paradox (97.3% claimed to avoid salt vs. 92.5% still adding it) could reflect social desirability bias in reporting, unrecognised salt in processed foods or culturally ingrained food preparation practices, as previously observed.³⁸ Critically low renal monitoring rates (15.2%) reflect systemic barriers, including equipment shortages,² the absence of national CKD screening policies¹ and financial constraints.²⁸

The results also revealed significant gaps in self-care practices among high-risk patients in the Buea Health District. While dietary practices showed relatively good adherence (97.3% reported avoiding high-salt foods and 88.5% consumed vegetables regularly), concerning deficiencies were noted in other critical areas. The finding that only 56.5% engaged in regular exercise, with 41.8% never exercising, is particularly alarming given the well-established benefits of physical activity for diabetes and hypertension management.⁴³ This prevalence of physical inactivity exceeds rates reported in similar African populations,⁴⁴ suggesting cultural or environmental barriers specific to this region.

The paradox between reported salt avoidance (97.3%) and actual salt addition (92.5%) likely reflects several factors, which may include underestimation of salt content in processed foods, cultural food preparation practices and social desirability bias in self-reported data.^38,45,46

The strong association between home glucose monitoring and better outcomes (χ² = 8.64, p = 0.003) supports findings from the Steno-2 study that intensive monitoring improves prognosis.⁴⁷ However, the low overall monitoring rates (29.8% for blood pressure, 15.2% for renal function) highlight systemic barriers, including the cost of monitoring equipment, limited health literacy and healthcare provider inertia.^28,31,48

The protective effect of regular exercise against hypertension (AOR: 0.07, p = 0.015) aligns with global evidence,²⁷ yet the lack of association with diabetes outcomes suggests that exercise intensity or duration may be insufficient for glycaemic control in this population.

Hypertension’s strong independent association with CKD (AOR: 4.28) mirrors findings from the Chronic Renal Insufficiency Cohort (CRIC) study.⁴⁹ The non-significant diabetes-CKD association (AOR: 0.32) contrasts with global data,¹ possibly because of shorter diabetes duration in our cohort, survival bias (early mortality in poorly controlled diabetics) and inadequate diagnostic sensitivity for diabetic nephropathy.

The age interaction findings (protective effect in the 73–83 years-old age group for diabetes, AOR: –0.17, p = 0.035) may reflect selective survival of healthier elderly individuals, cohort effects in disease management and diagnostic detection bias.³⁶

The consistent protective effect of tertiary education across all interaction plots supports the fundamental cause theory of health disparities.⁵⁰ Gender differences in disease patterns (higher baseline levels but steeper declines in females) may reflect biological risk factors,⁵¹ healthcare utilisation patterns²⁵ and social determinants of health.⁵²

Community-based group exercise programmes can improve chronic disease management. Subsidising home blood pressure and glucose monitoring devices enhances patient self-care. Clear salt labelling on packaged foods supports healthier dietary choices. Visual educational tools can improve health literacy among low-literacy populations. Training nurses in point-of-care CKD screening strengthens early detection capacity. These policies collectively support a proactive, equitable approach to preventing and managing hypertension, diabetes and CKD, particularly in resource-limited settings.