Assessment of non-communicable diseases screening practices among university lecturers in Ghana – a cross sectional single centre study

Study setting

The study was carried out at Kwame Nkrumah University of Science and Technology (KNUST), Kumasi between February to August, 2022. The study covered all the six (6) Colleges in the University.

Study design

This was a cross sectional study to ascertain health check practices among university lecturers. The study therefore captured the sex of lecturers who self-reported they were males or females. Therefore we did not inquire into the socially constructed roles, behaviors, expressions and identities of individual participants. The implication is that, no external or internal examination of body characteristics, genetic testing, or other means were conducted on study participants.

Study approach

The study employed a quantitative approach in which data was collected using questionnaires with both closed- and open-ended questions.

Study population

The study population involved 838 Lecturers across the six Colleges at Kwame Nkrumah University of Science and Technology (KNUST), Kumasi. A study of the lecturer population per college revealed that Colleges of Health Sciences (highest) and Agricultural/Natural resources (lowest) were the outliers (Quality Assurance and Planning Office, 2020).

Sampling technique

A simple probability technique was used to select the name of a college and the day/date to visit. Two sets of papers were folded with the names of colleges (set 1) and the day/date of the visit (set 2). A picker picked one folded piece of paper from each set and the name of the college and the day/date to visit were matched. In this case, the ordering of date and visit gave 1^st, College of Humanities & Social Sciences (COHSS), 2^nd College of Agriculture and Natural Resources (CANR), 3^rd College of Art & Built Environment (CABE), 4^th College of Engineering (COE), 5^th College of Science (COS) and 6^th College of Health Sciences (COHS). We then used the ‘walk-in’ system to select the study participants. By ‘walk-in’, the date and time of the visit to the Colleges were not announced to avoid possible biases in terms of lecturer-modifiable lifestyle behaviours. Within the days of visiting a college, any lecturer we met in his/her office was a potential study participant. All lecturers who consented to participate in the study were assessed instantly in their offices face-to-face (see details under data collection procedure and tools).

Sample size calculation

The sample size was obtained using Yamane’s (1967) formulae as shown below:

Where:

*n = is the population of Lecturers in at KNUST

*e = is the level of precision

Therefore:

n = 270 participants.

However, due to logistical constraints, 205 participants were contacted across the 6 Colleges at Kwame Nkrumah University of Science and Technology. The logistical constraints centred on consumables for the biochemical parameters and anthropometric measurements assessment; gloves, sanitizers, methylated spirits, glucometer strips, cotton wool, batteries (BP apparatus, weighing scales, glucometer gadgets) and others. We then applied simple proportions to get the number of lecturers to be consulted in each College.

Inclusion and exclusions criteria

Inclusion criteria were made up of all Lecturers on the KNUST campus who are in active service and consented to participate. All other staff not within this category were excluded from this research.

Ethical considerations

Ethical approval was sought from the CHRPE, KNUST with approval reference no: CHRPE/AP/581/21 granted on 8^th December, 2021. The aim of the research was explained to the participants. Those who consented to participate in the research were given consent forms to sign and date. Again, participants were assured of confidentiality. Participants were told that they were free to withdraw from the study with time. In other words, study participants were not coerced into the study.

Data collection procedures and tool

Data was collected using standardized structured questionnaires formulated for the current study based on the guidelines of Boynton and Greenhalgh (2004). The questionnaires were pretested at Akenteng Appiah-Menka University of Skills Training and Entrepreneurial Development in Kumasi, Ghana. Questions found to be uncomfortable to respondents or biased were amended and incorporated into the final set of questionnaires (see Extended data, Brenyah et al., 2023b). The questionnaires covered the socio-demographic characteristics of respondents, dietary intake, alcohol intake, issues with physical inactivity and tobacco use. Aside from these four main risk factors of NCDs, the questionnaire also captured the frequency of blood pressure checks, blood pressure outcome anytime it is checked (systolic and diastolic), frequency of general body check-ups, frequency of anthropometric measurement checks (weight and height), an assessment of impressions about the outcome of weight and height checks, an assessment of intended measures to be taken depending on the outcomes of weight and health checked. The weight, height and blood pressure were checked 3 times to ensure accuracy. Lecturers were also asked to identify if the content of their jobs has negative effects on their health status. With the assistance of two (2) State Registered Nurses, the research team and research assistants, lecturers were assessed on blood sugar status using glucometers, blood pressure using a digital sphygmomanometer, and weight and height using a weighing scale and stadiometer. Questionnaires were administered using Google Forms and submitted before the research team left the lecturer’s office. All these assessments were done in the offices of the lecturers. Instant results of blood sugar status, body mass index and blood pressure were released and interpreted for the respondents. All lecturers who were implicated in the BMI and blood pressure measurements were advised to visit the university hospital for further checks and treatment. Again, lecturers with poor dietary behaviour were also advised to visit a dietician at the university hospital.

Data management

Only the Research Team had access to data. Data was kept confidential. The researchers had planned of disposing data from the storage 5 years after the publication of this research. Collected data was entered and cleaned using a Microsoft Excel spreadsheet, and then imported into STATA version 14.0 (Stata Corp LP, College Station, Texas, USA) for statistical analysis and results presentation.

Data analysis

Descriptive statistics were used to summarize the characteristics of the study population by employing frequencies and percentages for categorical data. In addition, the association between categorical variables were assessed using Chi-square (χ²) or Fisher’s exact tests where appropriate with a p ≤ 0.05 assumed to be statistically significant. Both bivariate and multivariate logistic regression analyses were performed and adjusted for colleges’ effect to identify associations among the variables of interest. Variables having significant association in the logistic regression models were set at p ≤ 0.05 with a 95% confidence interval (95% CI) for both odd and adjusted odds ratios (OR, AOR).

Variables

Blood pressure (BP) was selected as the dependent variable, and in turn defined as Normal: ≤ 120/80 mmHg; Elevated: Systolic between 120-129 and diastolic ≤ 80 mmHg; Hypertensive: Systolic ≥ 130 or diastolic ≥ 80 mmHg. Then dichotomized into Normal blood pressure: ≤ 120/80 mmHg and high blood pressure (Hypertension): ≥ 130/90 mmHg for logistic regression analyses. Independent variables were socio-demographic characteristics; sex status (male/female), age, marital status, staff rank and lecturer’s colleges (categorized into binary variables; COHS/COS/COE and CABE/CANR/COHSS), family history of NCDs and health check status. In this study, the variable “very often” denotes (doing the activity in question more than 4 times a month), “often” denotes (doing the activity in question at least twice a month), and “not often” denotes (doing the activity in question once a month).

Socio-demographic characteristics of university lecturers

The study involved 205 participants including 176 (85.9%) men and 29 (14.15%) women based on their self-declaration of identity (Brenyah et al., 2023b). The mean age of participants was 46.3 ± 9.4 SD years with 39.02% aged 50 years and above as shown together with other variables in Table 1. The only disaggregated variables relating to sex (male and female) were the number and sex identity of study participants per their reported responses and the male/female risk ratio of developing hypertension.

Family history of non-communicable diseases among university lecturers

Participants with a self-reported NCD were 28.2%. The highest prevalence of the NCDs indicated was hypertension (66.8%) among the lecturers. Additionally, 50.1% of the Lecturers reported a family history of NCD, with hypertension accounting for 70.7% as shown on Table 2.

Distribution of health checks status by blood pressure among university lecturers

Table 3 shows that among the lecturers with hypertension, 59 (50.86%) often check their BP each month, whereas 22 (18.97%) reported not doing so often. Thus, there is a statistical association (p = 0.049) between the frequency of monthly BP checks and blood pressure. However, no associations were found between BP and the nature of BP results (p = 0.475), frequency of medical check-ups (p = 0.737) or BMI (p = 0.094).

Distribution of health checks status by blood sugar status among lecturers

The study found that 68.3% of lecturers undertake health checks by doing a blood sugar test twice a month. Again, 11.7% do the blood sugar test more than 3 times a month. Out of the 164 (80%) that took the test, 107 (65.3%) used the fasting blood sugar (FBS) and 57 (34.7%) used the Random Blood Sugar (RBS). Among those who took the test, 78 (47.5%) were not happy with their blood sugar status. The study found that, 61 (78.2%) claim their blood sugar status is mostly high (100 to 125 mg/dL (5.6 to 6.9 mmol/L) signifying prediabetes as shown in Table 4.

Factors associated with BP among KNUST lecturers

Table 5 shows bivariate and multivariate analyses for BP on socio-demographic characteristics and health check status. Lecturers in the age bracket 40 to 49 years were found to be associated (p = 0.036) with BP in the bivariate analysis, but not in the multivariate analysis (p = 0.114). Additionally, in the bivariate analyses, female lecturers were found to have a higher risk (OR 1.35; 95% CI 0.29-6.21) of developing BP (hypertension) compared to male lecturers. Lecturers in the colleges; COHS/COS/COE were found to be significantly associated (p = 0.016) with BP and were four times more likely (OR: 4.11, 95% CI: 1.30-12.95) to develop BP than those in the CABE/CANR/COHSS. Furthermore, it was shown that the frequency of lecturers’ BP checks (with the category ‘Often’) was significantly associated (p = 0.045) with their BP status. Although there was no association between blood pressure and any of the factors in the multivariate model, the analysis also revealed that lecturers who rarely check their weight (check monthly/yearly) were 91% more likely (AOR: 1.91, 95% CI: 0.09-41.98) to have BP than those who frequently check. Moreover, those with higher BMI (obese) had increased odds of BP that were more than six folds (AOR: 6.89, 95% CI: 0.61-77.9) compared with those with healthy weight. However, there were greater uncertainties surrounding the estimates of these odds.

Factors associated with BP among lecturers adjusting for colleges’ effect

The study also assessed if the variables assume different statistically significant associations if the logistics regression analyses were adjusted for effects from the colleges. It was noticed in Table 6 that strong statistical significance was found between BP checks and age (“40-49” p < 0.001, “50+” p = 0.05) and also with BMI (“overweight” p < 0.001, “obese” p < 0.001) in the bivariate analysis. The multivariate analysis showed several statistical associations were found between BP and its predicting factors including age: 40–49 (p < 0.001), the frequency of lecturers’ BP checks (with the categories “not often” p < 0.001, “Often” p < 0.001), job influence on lecturers BP (p = 0.037), the frequency of weight checks (with the category “never checked” p < 0.001) and BMI (“obese” p < 0.001) as shown on Table 6.

Limitations of the study

The study is limited in terms of scope. This relates to the fact that it was a single-centre study focusing on Kwame Nkrumah University of Science and Technology. Again, the limitation in scope also relates to the sample size. It would have been appropriate to use at least half of the lecturers’ population in the university. Despite these limitations, the outcome of the study is relevant and may serve as a baseline study for future research in this area.