Glycemic control and determinants among type 2 diabetes mellitus in a regional hospital in South West Region, Cameroon

Abstract

BACKGROUND

The global prevalence of diabetes among adults aged 29-79 years was found to be 10.5%. It is a global public health threat with a rising trend in morbidity and mortality. Poor glycemic control (GC) among patients with type 2 diabetes mellitus (T2DM) is a major determinant of diabetes-related complications. There are limited data on GC and associated factors among patients with T2DM in South West Region, Cameroon.

AIM

To assess GC and identify contributing factors among patients with T2DM in a regional hospital in South West Region, Cameroon.

METHODS

A cross-sectional study was conducted from February 2022 to July 2022 among 131 participants in Limbe Regional Hospital who were selected by convenience. Glycated hemoglobin (HbA1c) was measured by ion-exchange chromatography. Sociodemographic, clinical, and lifestyle data were collected, entered into Excel, and exported to Statistical Package for Social Sciences version 22 for analysis. A multivariate logistic regression analysis was conducted to assess the association between explanatory variables and GC. The level of significance was set at P < 0.05.

RESULTS

The mean age was 56 ± 5.1 years. Eighty-eight (67.2%) patients were female. The mean HbA1c was 8.8% ± 1.8%. Poor GC (HbA1c ≥ 7%) was registered in 106 (80.9%; 95% confidence interval: 73.1%-87.3%) participants. Lack of self-monitoring of blood glucose at home was associated with poor GC (adjusted odds ratio: 3.858, 95% confidence interval: 1.262-11.800; P = 0.018).

CONCLUSION

The majority of patients with T2DM had poor GC. Absence of self-monitoring of blood glucose at home was the main contributing factor for poor GC.

Key Words: Glycemic control; Determinants; Type 2 diabetes mellitus; Regional hospital; South West Region, Cameroon

Core Tip: Data on glycemic control (GC) in developing countries, particularly South West Region, Cameroon, are scarce. GC was assessed by measuring glycated hemoglobin and fasting plasma glucose recommended for settings limited in resources. Poor GC was recorded in more than half of the patients with type 2 diabetes mellitus. The findings from this study support clinical practice guidelines recommending the reinforcement of self-monitoring of blood glucose to avoid or lessen the complications of suboptimal GC. Future case-control and longitudinal studies are needed.

INTRODUCTION

Diabetes is a chronic metabolic disease characterized by a high level of blood glucose, which causes serious damage to the heart, eyes, kidneys, nerves, and blood vessels if poorly managed[1]. Diabetes mellitus is derived from the Greek word “diabaino”, which means siphon (to pass through), and from the Latin word “mellitus”, meaning honeyed or sweet. The moniker is fitting because in diabetes excess glucose is found in the blood and urine[2]. The American Diabetes Association (ADA) categorizes diabetes mellitus as type 1, type 2, gestational, and other specific types[3]. Type 1 diabetes mellitus is due to inadequate or lack of insulin, and type 2 diabetes mellitus (T2DM) is due to insulin resistance.

Glycemic control (GC) refers to the maintenance of blood glucose concentration in a patient with diabetes[4]. Glycated hemoglobin (HbA1c), postprandial glucose, and fasting plasma glucose (FPG) are markers of GC in patients with T2DM, but HbA1c is the gold standard of GC estimation because it gives the average blood glucose level over the past 3-4 months and indicates whether the treatment plan is working. Higher HbA1c values give an indication of the risk of developing diabetes complications. Good GC is defined as HbA1c less than or equal to 7% according to the ADA and HbA1c less than or equal to 6.5% according to the American College of Endocrinologists[4]. It is also described as FPG of 70-130 mg/dL (3.9-7.2 mmol/L), 110 mg/dL (6.1 mmol/L), and 100 mg/dL (5.5 mmol/L) by the ADA, the American College of Endocrinologists, and the International Diabetes Federation, respectively. Inadequate GC can substantially reduce patient quality of life and life expectancy and increase the healthcare costs of the disease via the complications of diabetes mellitus that can be microvascular (retinopathy, nephropathy, neuropathy)[4] or macrovascular (atherosclerosis, peripheral artery disease, cerebrovascular disease).

The global prevalence of diabetes was estimated by the International Diabetes Federation to be 10.5% (537 million) in 2021 among individuals aged 20-79 years. This prevalence has been estimated to rise to 11.3% (643 million) by 2030 and to 12.2% (783 million) by 2045. Globally, the prevalence of diabetes among 536.6 million people worldwide is higher (11.1%) in high-income countries than in middle-income (10.8%) and low-income (5.5%) countries. In Africa, roughly 4.5% (24 million) of people are living with diabetes mellitus[5], and still the African region bears the highest proportion of individuals living with undiagnosed diabetes. In Cameroon alone the age standardized prevalence of diabetes among individuals 20-79 years was reported to be 6.9% in 2024[6].

Studies in Iran and India reported poor GC among study participants (63.7% and 52.4%, respectively)[7,8]. The majority of the studies from Africa have currently reported proportions of poor GC among patients with T2DM ranging from 45.2% to 93.0%. A study in Egypt revealed a remarkably high prevalence (93%) of poor GC; meanwhile, a study in Ethiopia found a remarkably lower prevalence (45.2%) of poor GC[9,10]. Another five studies from Ethiopia have found higher poor GC proportions, from 64.1% to 73.8%[11-15]. Similarly, a current meta-analysis of the proportions of poor GC from studies conducted in Ethiopia demonstrated an elevated prevalence (61.92%) of poor GC in diabetes mellitus[16]. The most frequently reported determinants of poor GC include longer duration of diabetes mellitus, low level of education, overweight status, older age, non-adherence to diabetes treatment, non-adherence to diet, and lack of physical exercise[9-16].

The aforementioned evidence on GC in Ethiopia and other countries in Africa indicates an overall high prevalence of poor GC but with conflicting results. Data on GC and its contributing factors in Cameroon are scarce. There are also conflicting results on the determinants of GC among patients with diabetes mellitus. Therefore, the aim of this study was to assess GC and identify risk factors of poor GC among patients with T2DM attending the Limbe Regional Hospital in Cameroon.

MATERIALS AND METHODS

Study design, period, setting, and sampling

A hospital-based cross-sectional study was conducted among patients with T2DM for 6 months from February 2022 to July 2022 at the outpatient diabetic clinic of Limbe Regional Hospital located at Mile 1 in Limbe. Limbe Regional Hospital was considered most appropriate for this study because it caters to a wide range of patients from neighboring regions and rural areas. There were 203 registered patients with diabetes in this hospital at the time of this study. The literature review was conducted by searching the keywords of “GC”, “glucose control”, “determinants of GC”, and “associated factors of GC” in PubMed and Google Scholar databases.

Participants were enrolled into the study by convenient sampling. A sample size of 84 was determined using the Lorentz formula: n = Z²pq/d². The sample size calculation was based on a systematic review and meta-analysis study on the prevalence of prediabetes and diabetes mellitus among adults residing in Cameroon in which 5.8% was recorded as the overall prevalence of diabetes mellitus in a pooled sample of 37147 participants[17]. This gave a sample size of 84; however, a total of 131 participants were enrolled in this study because a larger, more diverse sample can better reflect the true variation in the target population, improving external validity or generalizability of the findings.

Study population

The study population comprised all patients with T2DM visiting the hospital for follow-up care. The management of patients at the diabetic center of the hospital involved a multidisciplinary team of a diabetologist, medical doctors, nurses, clinical pharmacists, and laboratory scientists.

Ethical considerations

Ethical clearance was obtained from the Faculty of Health Science Institutional Review Board of the University of Buea, No. 2022/1671-02/UB/SG/IRB/FHS, and written consent was provided by each participant. Administrative authorizations were obtained from the South West Regional Delegate of Public Health and the director of Limbe Regional Hospital.

Participants’ eligibility criteria and study variables

Patients with confirmed T2DM with at least 3 consecutive months of follow-up visits at the diabetic center were eligible for study inclusion. However, participants taking investigational or non-registered drugs or vaccines, receiving blood transfusion, with anemia or conditions that affect erythrocyte production, and who were critically ill or pregnant were disallowed study participation. Also, patients with neurological disorders or seizures were excluded for methodological and clinical reasons, as these conditions can impair cognition, memory, and comprehension, limiting participants’ ability to provide accurate self-reported data on lifestyle, adherence, and GC. Neurological disorders and their treatments can also independently influence GC and introduce confounding effects that could further bias study findings. Each of the final study participants was assigned a unique identifier to ensure that only one entry per participant was included in the analysis.

The level of GC measured by an HbA1c test was the dependent variable and the independent variables were sociodemographic, clinical, and behavioral factors. The sociodemographic factors included sex, age, occupation, monthly income, educational level, and marital status. The clinical factors included family history of diabetes, type of diabetic medication, body mass index (BMI), and duration of diabetes. The behavioral factors were self-monitoring of blood glucose (SMBG) at home, adherence to diet, adherence to medication(s), adherence to exercise, and smoking.

Data collection

After completion of a physician’s office visit, each participant was administered a structured and pretested questionnaire in a face-to-face interview to capture their sociodemographic, behavioral, and clinical data. The sociodemographic factors (age, sex, and level of education), anthropometric factor (BMI), clinical data (duration of diabetes, type of diabetic medication used), and lifestyle factor (type of exercise) were considered as factors affecting GC.

In order to prevent recall bias in regards to the participants’ reports of clinical data and to improve the identification of the confounders of the HbA1c test, we reviewed each patient’s medical records as well. The body weight of each participant was measured using a Kinlee calibrated weighing scale (Zhongshan Jinli Electronic Weighing Equipment Co., Ltd, Guangdong, China) with the participant wearing lightweight clothing and with shoes off. Height was measured using a stadiometer and recorded to the nearest 0.1 cm. BMI was obtained by dividing the weight (in kg) of the participant over the height squared (in m²). Duration of the participant’s T2DM and type(s) of diabetic medication used were determined by reviewing the participants’ medical records.

Adherence to the recommended diet was assessed using a food frequency questionnaire, which may have been subject to recall bias as it was self-reported by the participants. Dietary adherence was deemed as adequate or inadequate if the participant had followed the recommended diet for > 3 days or < 3 days, respectively, in a week[18]. Adherence to medication and exercise regimens were assessed using Morisky Medication Adherence Scale and World Health Organization guidelines on physical activity and sedentary behavior[19,20], respectively. Participants with Morisky Medication Adherence Scale-8 score of ≤ 6 were classified as low or non-adherent to medication(s). The participants were divided into those who used glucometers at home and those who did not use them. Participants were said to have made good use of the glucose meter if they used it either more than once a day, daily, or 2-4 times per week. Those who indicated not to have conducted a home self-monitoring of glucose were categorized into the lack of glucose self-monitoring group.

Venous blood was collected after the participant had fasted for 8-10 hours overnight, and aseptic techniques were employed prior to blood collection. Four milliliters of venous blood was drawn from each patient (2 mL into a dipotassium ethylenediaminetetraacetic acid tube and 2 mL into a fluoride oxalate tube) for subsequent determination of HbA1c and FPG, respectively. Blood samples collected in the fluoride oxalate tubes were first centrifuged at 3500 rpm for 5 minutes to obtain plasma for FPG measurement using the glucose oxidase method (Lot 4198; SGM New-Tem, Monza, Italy). Blood samples collected in the dipotassium ethylenediaminetetraacetic acid tubes were mixed well and then applied to an ion-exchange resin HbA1c testing kit (Lot 4409; SGM New-Tem, Monza, Italy). Biochemical tests were performed using a semi-automated chemistry analyzer (Cypress Diagnostics, Hulshout, Belgium). A quality control test was carried out for each batch of samples collected. All blood samples were analyzed on the same day of collection. The consistency and completeness of the study data was verified daily by the researchers working collaboratively as a team; proper coding and documentation of the data was also monitored and verified.

Operational definition of terms

The reference range of HbA1c was 4.0%-6.5%. Poor GC was defined as HbA1c ≥ 7%, and HbA1c < 7% was described as good GC according to recommendations by the ADA[21].

Adherence to an exercise regimen: Exercise was defined as getting 150 minutes/week of moderate aerobic exercise (e.g., brisk walking and stretching) or 75 minutes of vigorous exercise per week (e.g., jogging and cycling)[22].

Adherence to a diabetic medication regimen: The extent to which a person’s behavior in taking anti-diabetic medication(s) corresponded with agreed recommendations from a health care provider[8,22-24].

Adherence to a healthy diet: The degree to which the behavior of an individual regarding nutrient intake corresponded with the agreed recommendations by health care professionals. In adults, a healthy diet consists of 400 g (3-5 servings) of fruit and vegetables with sugars < 10% of total energy intake and fats < 30% of total energy intake in 1 day[25-29].

Statistical analysis

Data were entered into Microsoft Excel and exported to the SPSS version 22 (IBM Corp., Armonk, NY, United States). Sociodemographic and clinical data were reported as n (%). The association between sociodemographic characteristics and poor GC was assessed using logistic regression analysis. Only characteristics yielding a P value < 0.25 after a bivariate logistic regression analysis (i.e. age, BMI, adherence to healthy diet or eating plan, adherence to diabetic medication(s), duration of diabetes mellitus, and SMBG) were included in the subsequent multivariate logistic regression model. Of note, a P value cutoff of 0.05 after bivariate analysis may not be sufficient to identify variables that are known to be significant[30,31]. Univariate prescreening was the first strategy employed for reducing a large set of candidate variables into a smaller set; with this method covariates that were significant at a specific threshold based on a univariate model are typically the only ones included. As such, we respected that the use of a threshold of P < 0.05 could have resulted in the exclusion of important adjustment variables from the model, due to random variation[32], and we instead used a less stringent threshold (i.e. P < 0.25). After the multivariate analysis, predictors of poor GC with P < 0.05 were considered statistically significant. The Hosmer-Lemeshow test was conducted (Table 1).

Table 1 Logistic regression model fit according to Hosmer-Lemeshow test.

Parameter	χ2	df	P value
Age	0.000	1	1.000
BMI	0.000	0
Adherence to diet or eating plan	0.000	0
Adherence to diabetic medication	0.000	0
Duration of diabetes mellitus	0.000	0
SMBG	0.000	0

BMI: Body mass index; df: Degrees of freedom; SMBG: Self-monitoring of blood glucose.

RESULTS

Sociodemographic characteristics of study participants

One hundred and thirty-one patients with T2DM were enrolled for this study, 88 (67.2%) of whom were female. The mean age of the respondents was 56 ± 5.1 years. The majority of the patients were married (69.5%) and self-employed (73.3%), whereas 9.9% of patients were either private or public sector. Most (36.6%) of the participants had a secondary level of education (Table 2).

Table 2 Sociodemographic characteristics of patients with type 2 diabetes mellitus attending the Limbe Regional Hospital.

Parameter	Category	n (%)
Age in years	30-39	9 (6.9)
	40-49	33 (25.2)
	50-59	30 (22.9)
	≥ 60	59 (45.0)
	Total	131 (100)
Marital status	Single	23 (17.6)
	Married	91 (69.5)
	Divorce	5 (3.8)
	Widow/widower	12 (9.1)
	Total	131 (100)
Sex	Female	88 (67.2)
	Male	43 (32.8)
	Total	131 (100)
Occupation	Formal	13 (9.9)
	Retired	18 (13.7)
	Self-employed	96 (73.3)
	Unemployed	4 (3.1)
	Total	131 (100)
Level of education	Primary	48 (36.6)
	Secondary	58 (44.3)
	Tertiary	17 (13.0)
	Informal	8 (6.1)
	Total	131 (100)
Estimated monthly income in CFA franc	5001-10000	1 (0.8)
	10001-15000	9 (6.8)
	15001-20000	15 (11.5)
	20001-25000	79 (60.3)
	> 25000	27 (20.6)

CFA: Communaute Financiere Africaine.

Clinical data of study participants

The mean duration of T2DM since diagnosis was 8.11 ± 5.20 years. Forty-four (33.6%) of the respondents had a family history of diabetes. Sixty (45.5%) of them had lived with diabetes for at least 8 years. Obese (BMI ≥ 30 kg/m²) and overweight (BMI 25.0-29.9 kg/m²) status were recorded in 45 (34.4%) and 62 (47.3%) of the participants, respectively. One hundred and thirteen (86.3%) of the participants did not adhere to an exercise regimen and 68 (51.9%) of them did not adhere to the recommended diet/eating plan. More than half (77%) of the participants did not perform SMBG at home. One hundred and twenty-eight (97.7%) respondents were non-smokers, 106 (80.9%) of them were taking metformin and 20 (15.2%) of them were taking metformin with insulin or glimepiride (Table 3).

Table 3 Clinical characteristics of patients with type 2 diabetics attending the Limbe Regional Hospital.

Parameter	Category	n (%)
Family history of diabetes mellitus	Yes	44 (33.6)
	No	87 (66.4)
	Total	131 (100)
BMI, kg/m2	18.5-24.9 (normal weight)	24 (18.3)
	25-29.9 (overweight)	62 (47.3)
	≥ 30 (obesity)	45 (34.4)
	Total	131 (100)
Duration of diabetes mellitus, years	1-7	71 (54.2)
	8-14	44 (33.6)
	15-22	16 (12.2)
	Total	131 (100)
Type of diabetic medication(s)	Metformin	106 (80.9)
	Metformin + actrapid	1 (0.8)
	Metformin + mixtard	3 (2.3)
	Metformin + daonil	1 (0.8)
	Metformin + insulin or glimepiride	20 (15.2)
	Total	131 (100)
Adherence to exercise	Yes (150 minutes of moderate intensity/week or 75 minutes vigorous intensity/week)	18 (13.7)
	No (< 150 minutes of moderate intensity/week or < 75 minutes vigorous intensity/week)	113 (86.3)
	Total	131 (100)
Adherence to diet/eating plan	Yes (> 3 days per week)	63 (48.1)
	No (< 3 days per week)	68 (51.9)
	Total	131 (100)
SMBG at home	Yes	30 (23.0)
	No	101 (77.0)
	Total	131 (100)
Smoking	Non-smoker	128 (97.7)
	Ex-smoker	3 (2.3)
	Current smoker	0 (0.0)
	Total	131 (100)

BMI: Body mass index; SMBG: Self-monitoring of blood glucose.

GC among study participants

The mean HbA1c of the study participants was 8.8% ± 1.8%. Poor GC (HbA1c ≥ 7%) was diagnosed in 106 (80.9%) patients (Figure 1).

Figure 1 Glycemic control among patients with type 2 diabetes mellitus attending the Limbe Regional Hospital.

Factors associated with poor GC among study participants

Bivariate logistic regression analysis revealed that the following factors were significantly associated with poor GC: Ages between 50-69 years [unadjusted odds ratio (UOR): 0.230, 95% confidence interval (CI): 0.074-0.711; P = 0.011] and ≥ 70 years (UOR: 0.313, 95%CI: 0.106-0.919; P = 0.034); non-adherence to diabetic medication(s) (UOR: 0.192, 95%CI: 0.043-0.864; P = 0.032); lack of SMBG (UOR: 0.278, 95%CI: 0.109-0.706; P = 0.007); and non-adherence to recommended a diet/eating plan (UOR: 0.171, 95%CI: 0.059-0.490; P = 0.001). After adjusting for all the predictors of poor GC in a multivariate logistic regression model, patients with T2DM who did not monitor their glucose levels at home had an approximately 4 times higher risk (adjusted odds ratio: 3.858, 95%CI: 1.262-11.800; P = 0.018) of being diagnosed with poor GC (Table 4).

Table 4 Factors associated with poor glycemic control among patients with type 2 diabetics attending the Limbe Regional Hospital.

Parameter		Poor GC	Good GC	Bivariate logistic regression		Multivariate logistic regression
				UOR (95%CI)	P value	AOR (95%CI)	P value
Age, years	30-49	28	14	1		1
	50-69	65	11	0.230 (0.074-0.711)	0.011	2.819 (0.742-10.718)	0.128
	≥ 70	13	0	0.313 (0.106-0.919)	0.034	2.543 (0.725-8.911)	0.145
	Total	106	25
Adherence to diabetic medication(s)	Yes	73	23	1		1
	No	33	2	0.192 (0.043-0.864)	0.032	3.327 (0.582-19.020)	0.117
	Total	106	25
Duration of diabetes mellitus, years	1-11	73	24	1
	12-22	33	1	0.092 (0.012-0.710)	0.022	6.527 (0.726-58.667)	0.094
	Total	106	25
SMBG	Yes	19	11	1		1
	No	87	14	0.278 (0.109-0.706)	0.007	3.858 (1.261-11.800)	0.018
	Total	106	25
Adherence to diet/eating plan	Yes	43	20	1		1
	No	63	5	0.171 (0.059-0.490)	0.001	3.437 (0.941-12.559)	0.062
	Total	106	25
BMI, kg/m2	18.5-24.9	17	7	1		1
	≥ 25	89	18	0.491 (0.178-1.356)	0.170	1.483 (0.421-5.228)	0.540
	Total	106	25

GC: Glycemic control; UOR: Unadjusted odds ratio; CI: Confidence interval; BMI: Body mass index; SMBG: Self-monitoring of blood glucose; AOR: Adjusted odds ratio.

DISCUSSION

This study was conducted to assess GC and identify risk factors of poor GC among patients with T2DM at the Limbe Regional Hospital in Cameroon. We found that more than three-quarters of the patients had poor GC. Patient failure to monitor blood glucose level at home was the only predisposing factor of poor GC after employing the adjusted logistic regression model.

GC is the main goal of T2DM management. Our results on the proportion of patients with poor GC (81.0%) was consistent with findings in Nigeria (83.3%)[33] and Kenya (81.6%)[34]. This alarming prevalence of poor GC could be due to the high rate of self-employment and primary or secondary level of education. On the contrary, our finding was higher than estimates reported for Ghana (70%)[35,36], Ethiopia (68.3%)[37-39], northeast Ethiopia (70.8%)[40], India (37.5%)[8,41], Saudi Arabia (74.9%)[42,43], Tanzania (49.8%)[44], South Africa[45], Democratic Republic of Congo[46], Rwanda[47], Canada[48], and Iran[7]. This discrepancy in findings may be due to disparity in sample sizes, poor lifestyle conditions, cutoffs applied for HbA1c, failure to adhere to regular follow-up at diabetes clinic, and quality of care given to patients with diabetes at each hospital. It could also be because some studies used FPG to assess GC level while other studies considered HbA1c. Of note, our results conflict with the findings of Mideksa et al[13] in Mekelle-Ethiopia or of Yosef et al[14] in East Ethiopia. Insulin sensitivity is known to decrease with increasing age. Moreover, the progressive and chronic nature of diabetes mellitus makes it difficult for patients to maintain good GC, as the dysfunction of the beta cells impairs their insulin secretory function.

Patients with diabetes must engage in a number of vital self-care practices, including regular exercise, healthy eating, SMBG, medication adherence, and smoking cessation. Each of these activities provide benefits of increased glucose control, fewer complications, and enhanced quality of life. Lack of SMBG at home increased the odds of being diagnosed with poor GC among the patients with T2DM in this study. Approximately half (45%) of our patients with T2DM were aged 60 years and beyond. The majority of them had a primary and secondary level of education and low monthly incomes. These factors may have influenced the willingness and purchasing power of the patients to own personal glucometers for use at home or have access to SMBG systems in the hospital.

Our report corroborates with some evidence from studies of T2DM by Mideksa et al[13] in Ethiopia and Mbanya et al[49] in Cameroon. The former reported a significant association between non-glucometer use and poor GC among patients with T2DM. Similarly, the latter reported a significant relationship between self-management and good GC among patients with T2DM treated with insulin alone. Another study conducted by Abebe et al[50] indicated that more than three-quarters of the participants with poor GC also had poor adherence to SMBG. Unfortunately, SMBG at home has largely not been considered in the analysis of predictors of poor GC in studies conducted on the risk factors of poor GC among patients with T2DM. Informed decisions about medication dosage and when to urgently meet a healthcare professional are made by patients with diabetes if they monitor their blood glucose levels. Lifestyle changes can also be achieved in conjunction with self-monitoring, as the results provide evidence of both the negative effects of lack of change and positive effects of efforts to change. As such, they encourage and bolster the achievement of meeting goals towards optimal GC in patients with T2DM.

Strengths and limitations

We used HbA1c to assess GC among our patients. Our results showed that over 81.0% of patients with T2DM had poor GC. The contributing factor to this high prevalence was not carrying out SMBG at home. Due to the cross-sectional nature of this study, the cause-effect relationship of the independent variables to the outcome variable could not be made. Our findings might have been limited by recall bias as patients with T2DM were required to fill out the questionnaire (self-reporting). The use of convenient sampling also introduces bias, and we were unable to control for all potential confounders which may have played a role (such as medication dosage). The generalizability of our findings is further limited by the unavoidable fact that similar studies have yet to be conducted in other parts of our country, which precluded our ability to validate externally. Finally, there were also more females than males in our study sample and wide CIs were obtained; of note, this feature yielded smaller numbers in subgroups and restricted analytical power. Thus, the findings of this study are not maximized to fully represent the true variation of GC within the target population and support for an ongoing effort in research across the country is necessary.

CONCLUSION

The overall prevalence of poor GC among patients with T2DM attending the Limbe Regional Hospital was high. Lack of SMBG at home was identified as a contributing factor of the observed poor GC rates. Despite the inherent limitations to our study’s design, the findings from it underscore the urgent need for improved access to SMBG systems for patients with T2DM. Future studies should explore the impact of self-management education and low cost SMBG systems on GC among this national population. SMBG programs for patients with T2DM should be reinforced at regular follow-up appointments in order to avoid or lessen complications of suboptimal GC.

ACKNOWLEDGEMENTS

We acknowledge our gratitude to each of the patients who participated in this study and to the staff of Limbe Regional Hospital and Kumba Regional Hospital Annex Laboratory for their technical support.