Socio-demographic and modifiable risk factors of diabetes and hypertension among resource constrained patients from rural areas in Mdantsane Township in South Africa.

BACKGROUND: Recently, developing countries have shown a dramatic increase in non-communicable diseases (NCDs). The burden of NCDs in South Africa has increased over the past years resulting in an estimated 37% of all- cause mortality and 16% of disability-adjusted life years. Currently, diabetes mellitus (DM) and hypertension (HTN) are the two most prevalent NCDs associated with the rapid increase in mortality.
OBJECTIVE: To demonstrate the socio-demographic and modifiable risk factors of diabetes mellitus (DM) and hypertension (HTN) among South African adults.
METHODS: A cross-sectional analytical study was conducted in the Cecilia Makiwane Hospital serving the residents of Mdantsane. Relevant socio-demographic data, anthropometric measurements, triplicate blood pressure, fasting blood glucose and lipogram analysis were obtained from 265 outpatients.
RESULTS: Multivariate anlysis shows that; salt intake, smoking, elevated triglycerides and decreased high-density lipoprotein levels were significantly associated with DM with adjusted odds ratio of 0.18 (p=0.002), 0.26 (p=0.048), 2.19 (p=0.006) and 0.38 (p=0.001), respectively. Overweight and obesity were significantly associated with hypertension with odds ratio of 0.03 (p=0.01) and 0.06 (p=0.006), respectively.
CONCLUSION: The burden of DM and HTN on society can be drastically reduced with simple lifestyle changes, development of preventative strategies, large-scale screening and better disease management in South Africa.

Introduction

Non-Communicable Diseases (NCDs) have currently been identified as the leading cause of death worldwide. In the past decade developing countries have shown a dramatic increase in NCDs 1, 2. The burden of NCDs in South Afica has increased over the past years resulted in an estimated 37% of all cause mortality and 16% of disability-adjusted life years 3–5. Currently, Diabetes Mellitus (DM) and hypertension (HTN) are the two most prevalent NCDs associated with the rapid increase in mortality 6–10.

DM is defined as a chronic health condition associated with elevated blood sugar levels 11; 12, whilst HTN is characterized by a systolic blood pressure ≥ 140 mmHg and a diastolic blood pressure ≥ 90 mmHg 2. DM often co-exists with HTN since they both share common disease mechanisms and, in some instances, the one condition exacerbates the other 13. Currently, 425 million people are diagnosed with diabetes, whilst it is estimated that over a billion people worldwide are affected with HTN 2, 14. Both diseases have strongly been associated with an increased risk of kidney failure, obesity, stroke, blindness, nerve damage and cardiovascular disease (CVD) 2, 15–18.

DM and HTN have been shown to have a major impact on public health funding consuming a significant proportion of public health spending 7. However, these are described as lifestyle diseases, thus they can be prevented or managed by drugs and lifestyle modification19–21. Modifiable risk factors associated with DM and HTN include: tobacco use, alcohol consumption, physical activity and unhealthy diets 22. An unhealthy diet is defined, according to Forouzanfar et al 22, as a diet which is high in sodium, low in vegetables, low in fruit, low in whole grains, low in nuts and seeds, and low in seafood omega-3. Low- and middle-income countries are the most affected by these risk factors 23, 24.

The African region has been identified to have the highest burden of DM and HTN 30; 14. In Africa, DM is estimated at 15.5 million adults aged between 20–79 years31 and HTN is estimated at 46% in adults >25 years 30. The ever-increasing numbers of individuals diagnosed with these diseases are of great concern across the world especially in middle- and low-income countries 30. The present study highlights the burden and associated risk factors of DM and HTN in Mdantsane, a resource constrained township of South Africa.

In South Africa, the burden of NCDs is predicted to increase substantially in the next decades if the necessary preventative measures are not taken 25. Furthermore, strategies need to be implemented to effectively manage these diseases 26. Currently, there are limited studies exploring the prevalence and risk factors of DM and HTN in South Africa, especially within the economically disadvantaged population. The aim of this study was to investigate the socio-demographic and modifiable risk factors of diabetes and hypertension in one of the rural areas in South Africa, with resource constrained patients. Modifiable risk factors found significantly associated with diabetes and/or hypertension could be used to promote health education as a primary intervention.

Methods

Study area and design

A cross-sectional descriptive study was conducted in the Cecilia Makiwane Hospital (Mdantsane, South Africa) from July 2017 – October 2017. Mdantsane is located in the Buffalo Municipality and is a low-income residential township with a population of approximately 150000 27. The objectives of the study were explained to all participants and each participant signed a consent form indicating voluntary participation in the study. Information sheets were provided in both English and IsiXhosa languages. Prior to sampling, participants underwent physical examination and medical history was recorded.

Study population and sampling

Inclusion criteria for participants in this study were individuals aged ≥18 years and have been diagnosed with hypertension and/or diabetes for more than a year prior to the study. Exclusion criteria included pregnant women, patients diagnosed with type 1 diabetes and acute illnesses. Age, sex, monthly income, level of education, lifestyle profile (i.e. physical activity, smoking status, alcohol and diet), and family history of disease prevalence were obtained through interview from all of the participants. The use of anti-hypertensive and antidiabetic medications along with the durations of disease(s) was obtained from the patients' medical records. Eligible participants (N=265) were recruited sequentially at the study setting over the study period.

Data collection

A trained research nurse conducted anthropometric measurements of: weight to the nearest 0.1 kg, height to the nearest of 0.1 cm using a stadiometer, waist circumference, hip circumference and upper-arm circumference was measured using a tape measure. Measurements were taken with all participants wearing minimal clothing and no shoes. Blood pressure (BP) was measured using a validated automated digital blood pressure monitor (Microlife® BP A100 Plus). BP was recorded in triplicate and the average was used for analysis. Blood glucose was measured using Accutrend ® test strips. Body Mass Index (BMI) for each patient was calculated as weight (kg) divided by height (m2) and was categorised based on WHO criteria: underweight (<18.5 kg/m2), normal weight (18.5–24.9 kg/ m2), overweight (25.0-29.9 kg/ m2) and obese (30 or greater kg/m2). Patients with systolic BP (SBP) of ≥ 140 mmHg and ≥ 90 diastolic BP (DBP) were identified as hypertensive and patients with systolic and diastolic BP below 140 mmHg and 90 mmHg respectively were identified as normotensive.

Laboratory assessment

Fasting venous blood was obtained for all patients. The lipid profile which includes: total cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL) and high-density lipoprotein (HDL) for each participant was categorized according to the guidelines of the Heart and Stroke Foundation of South Africa 5, 28. In addition to this, the glycosylated haemoglobin (HbA1c) was assayed from blood samples of diabetic participants 29. All blood samples were sent to the clinical laboratory centre, i.e. National Health Laboratory Services (NHLS) of Cecilia Makiwane hospital and the East London private hospital.

Statistical analysis

Statistical analysis was performed using Statistical Package for Social Science (SPSS) version 25 for Windows (SPSS Inc., Chicago, IL, USA). The clinical laboratory data and anthropometric measurements were expressed as mean (n) ± standard deviation (SD). Differences between groups were assessed using chi-square test for statistical significance. Risk factors associated with DM and HTN are presented as percentages with the odds ratios (ORs) and 95% confidence intervals (CIs). The p-value ≤0.05 were considered statistically significant.

Results

In the study cohort, a total of 265 outpatients (of which n=175 were female and n=90 were male) were interviewed during a 3-month study (Table 1). The mean ages of men and women were 59.96±11.19 and 61.32±11.60 years, respectively. Other demographic, anthropometric and clinical laboratory measurements of the study participants are indicated in (Table 1).

Table 1

Characteristics of the study subjects in Mdantsane, Eastern Cape (July 2018)

Parameter	Female (n=175)	Male (n=90)	Total (n=265)
Age (years)	59.96±11.19	61.32±11.6	60.42±11.32
Weight (Kg)	87.45±21.46	81.62±16.06	85.46±19.94
Height (cm)	159.93±6.22	168.10±11.66	162.72±9.30
BMI (Kg/m2)	34.18±8.27	29.77±12.93	32.68±10.29
HbA1c (%)*	10.40±2.80 (n=85)	10.48±3.91 (n=32)	10.42±3.12 (n=117)
FBG (mmol/l)*	12.65±5.11 (n=85)	13.11±3.92 (n=32)	12.78±4.80 (n=117)
Systolic blood pressure (mmHg)	155.67±20.53	157.04±21.65	156.14±20.88
Diastolic blood pressure (mmHg)	92.56±13.18	93.87±13.45	93.00±13.26
Heart rate (pbm)	84.14±13.60	80.01±14.43	82.74±13.99
TC (mmol/L)	5.02±1.27	4.53±1.14	4.86±1.25
HDL (mmol/L)	1.30±0.36	1.35±0.44	1.31±0.39
LDL (mmol/L)	2.63±1.14	2.35±0.98	2.54±1.09
TG (mmo/L)	1.71±1.01	1.88±1.06	1.77±1.03

BMI – Body Mass Index, HbA1c - glycated haemoglobin, FBG-Fasting blood glucose, TC – Total Cholesterol, HDL – high-density lipoproteins, LDL – Low Density Lipoproteins, TG – Triglycerides, CRT – Creatinine, GFR - glomerular filtration rate. n – Total number of samples/patients, P-value > 0.05.

HbA1c and RBG were only measured for patients diagnosed with DM thus n vary. Values are presented as means ± standard deviation

Table 2 and 3 indicates the socio-demographic and modifiable risk factors of non-diabetic and diabetic groups as well as among non-hypertensive and hypertensive groups. It is important to note that approximately 40% of the study cohort was co-morbid. In both NCDs, the proportion of females is higher than males, however, sex was only shown to be significantly associated amongst diabetic patients (p-value = 0.043). Amongst diabetic patients, smoking status; salt intake, TG and HDL were all significantly associated with disease incidence with p-values of 0.015; 0.004, 0.012 and 0.003 respectively (Table 2). All other factors, i.e. age, educational level, physical activity, alcohol consumption, TC and LDL were not significantly associated with DM (Table 2). BMI was the only modifiable risk factor that showed significant association amongst hypertensive patients with a p-value of <0.0001 (Table 3). Factors not significantly associated with HTN were: sex, age, educational level, smoking status, physical activity, salt intake, and alcohol consumption, TC, TG, LDL and HDL (Table 3).

Table 2

Socio-demographics and Modifiable risk factors among diabetes (n=265)

		Group
Variables	Subgroups	Non-diabetic		Diabetic		X2,P Value
		n = 148	%	n=117	%
Gender	Male	58	64.4	32	35.5
	Female	90	51.4	85	48.6	4.08, 0.040
Age	Less than 50 years	28	63.6	16	36.4
	More than 50 years	120	54.3	101	45.7	1.30, 0.260
Educational level	No formal education	12	60.0	8	40.0	4.78, 0.190
	Primary education	34	54.0	29	46.0
	Secondary education	97	58.1	70	41.9
	Higher education	4	28.6	10	71.4
Smoking status	Never smoked	99	52.1	91	47.9	8.45, 0.015
	Quit smoking	29	56.9	22	43.1
	Current smoker	20	83.3	4	16.7
Physical activity	More than 3 times / week	8	50.0	8	50.0	0.25, 0.880
	1–2 times/ week	119	54.6	92	43.4
	No physical activities	21	55.3	17	44.7
Salt intake	No salt intake	11	37.9	18	62.1	11.15, 0.004
	Normal salt intake	103	53.9	88	46.1
	Increased salt intake	34	75.6	11	24.4
Alcohol consumption	Never drank	78	54.9	64	45.1	3.63, 0.160
	Quit drinking	40	50.6	39	49.4
	Occasional drinker	30	68.2	14	31.8
BMI (Kg/m2)	<18.5	4	57.1	3	42.9	0.28, 0.960
	18.5–24.9	21	55.3	17	44.7
	25.0–29.9	37	53.6	32	46.4
	≥30	86	57.3	64	42.7
TC (mmol/L)	Increased	52	46.8	59	53.2
	Normal	96	56.8	58	43.2	0.15, 0.700
TG (mmol/L)	Increased	52	46.3	59	53.7
	Normal	95	62.3	58	37.7	6.28, 0.012
HDL (mmol/L)	Decreased	74	67.3	36	32.7
	Normal	69	48.6	73	51.4	8.81, 0.003
LDL (mmol/L)	Increased	50	56.8	38	43.2
	Normal	98	55.4	79	44.6	0.05, 0.820

TC= Total Cholesterol, TG= Triglyceride, HDL= High density lipoprotein, LDL= Low density lipoprotein, mmol= mill mole, L= litre. Location: Mdantsane, Eastern Cape (July 2018)

Table 3

Factors affecting the Modifiable risk factors of hypertension in study subjects (n=265)

Variables	Subgroups	Group				X2, P value
		Non-hypertensive		Hypertensive
		n = 13	%	n=252	%
Gender	Male	6	6.6	84	93.3	0.91, 0.34
	Female	7	4	168	96
Age	Less than 50 years	4	9.1	40	90.9	1.98, 0.16
	More than 50 years	9	4.1	212	95.9
Educational level	Uneducated	0	0	20	100	1.46, 0.69
	Primary	4	6.3	59	93.7
	Secondary	8	4.8	159	95.2
	High education	1	7.1	13	92.9
Smoking status	Never smokers	8	4.2	182	95.8	1.17, 0.56
	Quit smokers	4	7.8	47	92.2
	Current smokers	1	4.2	23	95.8
Physical activity	More than 3 times / week	0	11.1	16	88.9	0.88, 0.64
	1–2 times/ week	11	4.3	200	95.7
	No physical activities	2	5.3	36	94.7
Salt intake	No salt intake	0	0	29	100	1.84, 0.40
	Normal salt intake	10	5.2	181	94.8
	Increased salt intake	3	6.7	42	93.3
Alcohol consumption	Never drank	4	2.8	138	97.2	3.32, 0.19
	Quit drinking	5	6.3	74	93.7
	Occasional drinker	4	9.1	40	90.9
BMI (Kg/m2)	<18.5	2	28.6	5	71.4	21.34, <0.0001
	18.5–24.9	6	15.8	32	84.2
	25.0–29.9	1	1.4	68	98.6
	≥30	4	2.7	146	97.3
TC (mmol/L)	Increased	5	4.9	98	95.1
	Normal	8	4.9	154	95.1	0.001, 0.98
TG (mmol/L)	Increased	4	3.5	107	96.5
	Normal	9	5.8	145	94.2	0.69, 0.41
HDL (mmol/L)	Decreased	2	1.8	108	98.2
	Normal	8	5.6	134	94.4	2.37, 0.12
LDL (mmol/L)	Increased	4	4.5	84	95.5
	Normal	9	5.1	168	94.9	0.04, 0.85

TC= Total Cholesterol, TG= Triglyceride, HDL= High density lipoprotein, LDL= Low density lipoprotein, mmol= mill mole, L= litre. Location: Mdantsane, Eastern Cape (July 2018)

Table 4 and 5 describe the univariate and multivariate analyses for diabetes and hypertension. The univariate results show a risk association between diabetes and smoking status, salt intake, TG and HDL (Table 4). The results gave an indication that smoking status has an impact on diabetes using never smoked as the reference, those who quit smoking had significantly lesser odds (p-value = 0.824) of diabetes in comparison to those who are currently smokers (p-value = 0.048) (Table 4). Furthermore, an increased salt intake, increased TG and a decreased HDL-C also demonstrated significantly higher odds (p-value = 0.01) of diabetes. Gender was not significantly associated with diabetes (p-value = 0.210). In table 5, an increased BMI in comparison to the underweight (as per WHO standards) has higher odds of hypertension (p-value=0.012 and p-value= 0.006). In addition to this, participants who have normal BMI have lower odds of hypertension (p-value = 0.420).

Table 4

Univariate and Multivariate analysis for risk factors of Diabetes status

Factors	Diabetic N (%)	Non- Diabetic N (%)	Unadjusted Odds ratio	Adjusted odds ratio	P-value
Sex
Male	32 (35.6)	58 (64.4)	1	1
Female	90 (51.4)	85 (48.6)	1.71 (1.01–2.89)*	1.55 (0.78–3.1)	0.212
Salt intake
No salt intake	18 (62.1)	11 (37.9)	1	1	0.009
Normal salt intake	88 (46.1)	103 (53.9)	0.52 (0.23–1.17)	0.54 (0.19–1.08)	0.075
increased salt intake	11 (24.4)	34 (75.6)	0.2 (0.07–0.54)*	0.18 (0.06–0.55)*	0.002*
Smoking status
Never smoke	91 (47.9)	99 (52.1)	1	1
Quit smoking	22 (43.1)	29 (56.9)	0.83 (0.44–1.54)	1.09 (0.51–2.36)	0.824
Current smokers	4 (16.7)	20 (83.3)	0.22 (0.07–0.66)*	0.26 (0.07–0.98)	0.048*
TG- Cholesterol
Normal level	58 (37.7)	96 (62.3)	1	1
Increased level	59 (53.2)	52 (46.8)	1.88 (1.15–3.08)*	2.19 (1.3–3.8)	0.006*
HDL-Cholesterol
Normal level	73 (51.4)	69 (48.6)	1	1
Decreased level	36 (32.7)	74 (67.3)	1.06 (0.27–0.77)*	0.38 (0.22–0.67)	0.001*

P-value <0.05.

Location: Mdantsane, Eastern Cape (July 2018)

Table 5

Univariate and Multivariate analysis for risk factors of hypertension

Factors	Hypertensive N (%)	Non-hypertensive N (%)	Unadjusted Odds ratio	Adjusted odds ratio	P-value
BMI
<18.5	5 (71.4)	2 (28.6)	1	1
18.5–24.9	32 (84.2)	6 (15.8)	0.47 (0.07–3.0)	0.35 (0.048–2.56)	0.304
25.0–29.9	68 (98.6)	1 (1.4)	0.04 (0.003–0.48)*	0.03 (0.002–0.44)*	0.010*
≥30	146 (97.3)	4 (2.7)	0.07 (0.01–0.47)*	0.06 (0.007–0.447)*	0.006*
Smoking status
Never smoke	182 (95.8)	8 (4.2)	1	1
Quit smoking	47 (92.2)	4 (7.8)	1.93 (0.60–6.7)	0.764 (0.16–3.61)	0.700
Current smokers	23 (95.8)	1 (4.2)	0.99 (0.012–8.3)	0.34 (0.04–3.9)	0.400
Alcohol consumption
Never drank	138 (97.2)	4 (2.8)	1	1
Quit drinking	74 (93.7)	5 (6.3)	2.33 (0.61–8.95)	3.7 (0.67–20.6)	0.14
Occasional drinker	40 (90.9)	4 (9.1)	3.45 (0.83–14.4)	2.01 (0.42–9.67)	0.38

P-value <0.05

Location: Mdantsane, Eastern Cape (July 2018)

Multivariate logistic regression analysis showed that after adjusting for all significant factors, an increased salt intake and BMI were significantly associated with DM and HTN respectively (Table 4 and 5).

Discussion

South Africa has been reported to have the highest incidence of DM in the African continent 14. Amongst the modifiable risk factors, significant association was shown with tobacco intake, increased salt intake, TG and HDL. Tobacco smoking is well established as a risk factor for multiple diseases and has been associated with DM in multiple cohort studies 32–34. The present study showed that smoking was associated with the probability of developing DM. This finding is consistent with previous studies conducted in Korea 35; 36. Current smokers and ex-smokers display a greater probability of developing DM than non-smokers, however in this study, the increased risk of ex-smokers were not statistically significant. Previous studies conducted by Jee et al. 35 and Hur et al. 37 also reported the increased risk of ex-smokers as insignificant.

The WHO (2016) recommends that patients with DM should reduce their dietary salt intake 2. The precise relationship between dietary salt intake and DM is not well defined, however, excessive salt intake is well associated with hypertension and CVDs. In the present study, an increased salt intake was significantly associated with the higher incidence of DM. Previous studies also demonstrated an association between high dietary salt intake and DM 38–40. Increased TG levels have been associated with an increased risk of DM 41–44 and in this study cohort, similar results were observed.

In addition, this study also found that the odds of having DM were increased with a decrease in HDL. Similar findings have been reported in African 45–48, European 49 and United States communities 50; 51. Lower levels of HDL concentrations have been associated with many diseases such as CVDs 52–55, nephropathy 56 and coronary heart disease 57; 58. Although, levels of TC and LDL in diabetic individuals are reportedly comparable with that found in non-diabetics, low levels of HDL and elevated TG have been reported in T2DM patients as the probable cause of CVD 52–54. It has also been observed that HDL alone might not be a good indicator of increase DM risk since most of the subjects had lower total cholesterol. Moreover, lower levels of HDL in the present study might be because of the lower cholesterol.

A high BMI is a risk factor that is often associated with DM 59, however, in this study; it was significantly associated with HTN, since DM and HTN co-exist in approximately 40% of the study cohort, this could be an explanation for this observation. Furthermore, many studies suggest that a high BMI contributes to hypertension 60–64. It is well established that smoking increases the risk of hypertension; however, the significance of this association may differ between populations 65. In this study, no significant association was observed between hypertension and smoking status. These findings are contrary to other studies 66–69.

The following limitations need to be considered; the cross-sectional design and recruitment of participants from one study centre might limit the generalisation of the findings. In addition, the scope of this study needs more samples drawn from broader population across the country to conclude on an association that exists between DM and HTN with the discussed variables.

Conclusion

DM was associated with smoking and salt intake; whilst hypertension was associated with increasing BMI. Development of best practices for affordable and effective programs in screening, prevention, detection and treatment of DM and HTN is essential. In order to reduce the burden of NCDs, comprehensive intervention strategies should be implemented across the country. Future studies with larger sample size should be done to identify or generate local modifiable risk factors for the development of DM and HTN.