Anthropometric risk characteristics and blood pressure as predictors of pre-diabetes: A cross-sectional comparative study

Public health importance

Pre-diabetes and obesity constitute a major public health crisis. They are the key drivers of the epidemic associated with type 3 diabetes mellitus and cardiovascular diseases globally. Within our setting, individuals may be obese/pre-diabetic and may not be aware, as these are often asymptomatic. Besides being a precursor to diabetes mellitus, pre-diabetes is an independent risk factor for other chronic diseases; hence, its public awareness, early screening/detection, and reversal are key to sustaining a healthier society.

Study population

The study population consisted of all the adult patients aged 18 years and above who presented for primary health care during the study period.

Eligibility criteria

Study design

It is a cross-sectional comparative study.

Sample size determination

The sample size was determined using the formula for a comparative study with quantitative outcomes.^[15]

Where n = minimum sample size, r = ratio of controls (normoglycemics) to cases (pre-diabetics).

Here r = 1, because controls and cases were equal, standard deviation (SD) = Degree of variability of observations of fasting blood glucose (FBG) in the population (taken as 2.8),^[16] d = expected mean difference between FBG of case and control (d = 6.5−4.8 = 1.7 mmoL/L),^[17] Z_β and Z_α/2 are constants (0.84 and 1.96, respectively).

Substituting values, n = 42.5 = (43 participants/group).

To increase the power of the study, n was approximated to a minimum sample size of 50 for each.

Ethical considerations

An approval for the study was obtained from the Research and Ethics Committee of Alex Ekwueme Federal University Teaching Hospital, Abakaliki, with approval number: FETHA/RE/VOL.2/2018/078 and dated July 31^st, 2018.This research work complied with the Helsinki Declaration 2013 on human research. The nature and purpose of the study were explained to the respondents in a language they could understand. They were assured of confidentiality. Participation in the study was voluntary, and they were free to withdraw their consent at any point without any adverse consequences to them. The costs of the study investigations were borne by the researcher.

Sampling technique

The study and its purpose were explained to the participants, and written informed consents were obtained. There was no hospital record of pre-diabetic attendance; hence, consecutive methods were used to select participants.^[13] The patients’demographic details were taken in a separate booklet; there were no identifying details in the questionnaires.

Study instruments

Pre-tested interviewer-administered structured questionnaires that centered on sociodemographic/anthropometric characteristics and BP were used. Height and weight were measured to the nearest 0.1 kg and 0.1 cm, respectively, using a combined weight scale and stadiometer (SECA brand, made in Hamburg, Germany); the BMI was calculated as a ratio of weight (kg) to height² (m²). The BP was measured using a mercury sphygmomanometer and a stethoscope. A non-elastic dressmaker’s tape was used to measure the WC to the nearest 0.1 cm. Calibrated Accu-Chek Active glucometer and strips (by Roche Germany) were used in blood glucose measurement. The Accu-Chek glucometer had an accuracy level close to the colorimeter technique.^[18] All measurements were carried out using standard protocols.

Methods of data collection

One research assistant was trained in the standard procedures of the different measurements. The participants who had eaten or arrived late were given different appointment dates in batches of ten. This corresponded with their follow-up visit days. They had their FBG (mg/dL) tested after an overnight fasting of at least 8 h. The glycemic levels were categorized as normal (FBG of 65–109 mg/dL), impaired fasting glucose (FBG of 110–125 mg/dL), and diabetes (FBG of ≥126 mg/dL).^[2,18] Those with pre-diabetes were matched for age and sex and had their BP, BMI, and WC measured and recorded. The BMI was classified into two categories: Normal = 18.5–24.9 kg/m² and abnormal (overweight/obesity) ≥25.0kg/m² and analyzed. The WC was categorized as normal (<94 cm for males and <88 cm for females) and abnormal (≥94 cm for males and ≥88 cm for females), using the International Diabetes Foundation classification.^[19] Social class was classified into upper, middle, and lower classes based on wealth, occupation, and educational attainment.^[20] A structured, interviewer-administered questionnaire was used to collect the data. Participants who did not meet the inclusion criteria were dropped from the study.

Data analysis

Frequencies, percentages, means, and standard deviations were calculated for descriptive statistics. The data generated were analyzed using the Statistical Packages for the Social Sciences (SPSS) (IBM SPSS version 23). The outcome (dependent) variable was pre-diabetes. Independent variables were sociodemographic/anthropometric characteristics and BP. Chi-square tests were used to test the associations, and logistic regression was used to find variables that had an independent association with pre-diabetes. The level of significance was set at p < 0.05 and the confidence level at 95%.

Socio-demographic characteristics

Occupational status, educational attainment, and social class singly predicted pre-diabetes [Table 2]; while age, gender, marital status, and family history of diabetes did not predict pre-diabetes. Again, a community-based cross-sectional study by Endris et al., in Dessie city, Northeast Ethiopia, reported pre-diabetes to be more prevalent after the age of 60 years, while a study in Kenya reported that women, compared with men, were significantly more likely to be obese.^[21,22] Another study reported that women were more likely to visit their physicians than men for both physical and internal health enterprises.^[23] Other studies reported male preponderance in pre-diabetes.^[24,25] The independent association between advanced educational attainment and occupational status with pre-diabetes, as reported in this study, could be due to the link between the two socio-demographics linked to having little or no time for physical activity. Furthermore, higher income among the educated tends to give them access to a Western diet and lifestyle which they can afford, and these encourage the eating of fast or refined food, therefore prepping them to dysglycemia. A multi-center study noted that physical activity lowers the risk of type 2 diabetes, while another narrative review of changes in dietary habits noted that the westernization of the traditional African diet can lead to fat accumulation, obesity, glucose intolerance, and eventually diabetes.^[26,27] In Ghana, a cross-sectional study showed that the odds of diabetes increased with higher educational levels.^[25]

The odds of developing pre-diabetes were about 5 times higher among the unemployed. Unemployment could lead to stress, which triggers the stress hormones that affect glucose metabolism. Rautio et al., in a birth cohort study in Northern Finland, implicated unemployment as an independent risk factor for pre-diabetes.^[28] This was attributed to overactivity of the hypothalamic-pituitary-adrenal axis and cortisol product, as well as behavioral factors. Another cross-sectional study made the same assertion.^[29] The probability of developing pre-diabetes was about 3 times higher in the higher social class compared to the lower social class [Table 2]. The reason for this finding could be that persons in this class do not eat healthily (cannot afford and consume fast or refined foods) and live a sedentary lifestyle. Another study proved that higher wealth was significantly associated with a higher prevalence of pre-diabetes compared to lower wealth.^[9] Contrary to our finding, a study in Namibia reported pre-diabetes to be more among the lower social class than the higher class, and residents of households categorized as middle wealth index were associated with lower odds of pre-diabetes.^[8]

Anthropometric characteristics

There was a statistically significant association between the BMI and WC and pre-diabetes [Table 3]. The odds of developing pre-diabetes were about 2 times and 3 times increased in elevated BMI and WC, respectively [Table 4]. This finding is presumably because abnormal BMI is associated with increased obesity, causing inhibition of insulin receptor signaling. In a study by Rahmanian et al., the odds of being pre-diabetic were higher in overweight individuals.^[1] A population-based cross-sectional study in Turkey reported that abnormal BMI was independently associated with increased prevalence of abnormal glucose regulation.^[3] In central Saudi Arabia’s cross-sectional study in a selected military community, abnormal BMI significantly increased random blood glucose levels.^[4] Furthermore, in another study in Ghana, obesity was associated with insulin resistance but was not significantly associated with blood glucose levels among the study participants.^[25]

Likewise, the independent association between pre-diabetes and an abnormal WC could be from abnormal fat deposits, which play a significant part in the development of insulin resistance. Another study reported that abnormal WC significantly increased random blood glucose levels among a selected military community.^[4] A study in Spain asserted that abdominal obesity, among other factors, was independently associated with pre-diabetes.^[6] Again, general obesity, though a risk factor in both sexes, was not statistically significant among men after adjusting for all factors.

Co-morbid condition (BP)

There was a statistically significant difference in the mean SBP between the pre-diabetics and the normoglycemic participants. There was still no statistically significant difference in mean DBP [Table 5]. The reason for the elevation in SBP may presumably be because of the metabolic derangement that occurs in abnormal glucose regulation as part of the metabolic pattern, as seen in statistically significant advanced weight, BMI, and WC among the participants [Table 5]. Henry et al., reported that mean SBP was significantly associated with abnormal glucose regulation.^[30]

Study limitations

It was a hospital-based cross-sectional study; hence, the findings may not reflect exactly what happens in reality in the larger population. The diagnosis of pre-diabetes was based on impaired fasting glucose alone instead of a combined impaired glucose tolerance and glycated hemoglobin, which were costly, with a resultant lower prevalence of pre-diabetes. A number of patients may not have observed real fasting before the screening but may have been attracted to volunteer for screening, since it was not paid for; hence, in such cases, one may unknowingly be reporting or using RBS values for FBS.

Recommendations:

The use of a larger population size in future studies is advocated, as well as a community-based study with a larger sample size, the findings of which will reflect the exact reality. The impact of lifestyle on the development of pre-diabetes should be an area for further studies since it is a potential cofounder. We recommend the use of these variables in the development of clinical-based protocols in the screening and diagnosis of pre-diabetes, as well as their incorporation into primary care clinics as better settings for pre-diabetes screening as part of the ongoing reform of the health sector in Nigeria.