Age-group differences in risk perceptions of non-communicable diseases among adults in Diepsloot township, Johannesburg, South Africa: a cross-sectional study based on the health belief model.
While adults in their 70s and older age groups are generally at greatest risk of dying from chronic diseases such as CVDs, death rates among this age group have reversed and continue to decline compared with the age groups 25-39 years, 40-54 years, and 55-69 years, showing less progressive patterns in the decline of the annual NCD-related death rate.  It is presumed that high consumption of energy-dense foods, being overweight or obese, coupled with high rates of physical inactivity, have been accelerating this trend in the younger adult population.  This could possibly be due to how the young adult population perceive the risks of NCDs compared with the older adult population.
The Health Belief Model (HBM) is one of the most effective health behaviour-change models that comprises six constructs (perceived susceptibility, severity, benefits, barriers, cues to action and self-efficacy), which explains why an individual may or may not adopt preventive measures against the disease of interest--NCDs in this case.  Cross-sectional studies among Saudi female employees  and rural Turkish women  found that poor practice of monthly breast self-examination and mammography as preventive measures of breast cancer were not statistically significantly associated with age.
Furthermore, age did not predict women's attitudes towards this practice. Similar findings were reported in studies conducted in two villages near Kathmandu (Nepal)  and in Karachi (Pakistan)  in relation to their attitude towards cardiovascular health  and modifiable risk factors for heart diseases.  In contrast, the level of knowledge and practice of cardiovascular health were statistically different according to age,  where participants <35 years old were more likely to have a greater adequate knowledge and better practice than those who were >45 years of age. [12, 13] This was consistent with the results of other studies on knowledge of risk factors and prevention regarding diabetes in eastern Saudi Arabia,  and of chronic diseases in Jinan (China).  A cross-sectional study among black and Hispanic females in the USA found that younger-aged females (<55 years) were less conscious of heart disease being the leading cause of mortality in women and were less likely to state that they were very well or knowledgeable concerning heart disease compared with older-aged women.  Another study in the USA found that age was not a predictor of knowledge of perceived susceptibility and perceived self-efficacy to heart diseases among the African-American population. 
The role of age in the adoption of NCD preventive measures is contentious, as studies have disputed whether there is a positive or negative association between the two variables. [10,11,16,18,19] However, findings of these studies cannot be generalised to Diepsloot's population owing to dissimilar sociodemographic characteristics. Studies in African settings on this topic are scarce, with none conducted in SA, especially among informal settlement dwellers, who are already at high risk of developing NCDs because of their unhealthy lifestyle.  Health beliefs, attitudes and knowledge are key paradigms of health behaviour theories. In particular, perceptions of disease risk, control and severity are integrated into social cognitive models such as HBM, as they trigger health behaviours, mediate the impact of other risk factors, are flexible and responsive to change, and are targets for disease interventions.  Consequently, the development of effective interventions for people at increased risk of NCDs is dependent upon understanding their beliefs concerning NCDs. NCD-related DALYs are expected to rise by 27% across regions of Africa compared with 17% globally by 2020,[22,23] with more than half occurring prematurely (in persons <70 years of age). [22,24 In SA, adherence to the use of antiretroviral therapy (ART) has not only extended survival and ageing among HIV-infected individuals, but also supplemented the upsurge in NCD comorbidities among people living with HIV. [25,26] The prevalence of dysglycaemia was 2-4-fold greater among HIV-intected groups compared with that in a community-based survey.  The prevalence of multimorbidities (hypertension and diabetes mellitus) was greater among people living with tuberculosis (TB) and younger HIV-infected individuals on ART compared with those without TB and HIV infection.  A better understanding of the relationships between age groups and risk perceptions of NCDs is required. Our study is particularly important, as NCDs in SA occur simultaneously with an ageing HIV-positive population.  This has important implications, as it will not only contribute towards the burden of premature deaths in Africa, but also worsen social inequities and poverty in informal settlements in SA because most expenses for healthcare are private and out-of-pocket; such expenses weigh more profoundly on individuals least capable of affording them. The results of this study will assist to inform, rationalise and substantiate the necessity for evidence-based policy, practice and behavioural change interventions aimed at improving adoption of NCD preventive measures in informal settlements as an approach to decrease the global burden of NCDs in SA. The aim of this study was to: (z) examine the relationships between age groups and risk perceptions of NCDs; and (ii) assess the distribution of risk perceptions of NCDs in relation to age groups.
This was a cross-sectional design using secondary data obtained from Community AIDS Response (CARe) in 2014, assessing risk perceptions of NCDs based on HBM in Extension 2 (Blocks I, J, K and L) of Diepsloot township, Johannesburg, SA.
Diepsloot is a heavily crowded informal settlement in the extreme north of Johannesburg. [28,29] It is situated in the recently formed Region A (formerly known as Region 1 and 2).  It is currently residence to-360 000 individuals.  The majority of the inhabitants are low socioeconomic class blacks.  Nearly 73.7% of the 20-59-year age group are employable; [28-29] nonetheless, unemployment rates remain high owing to low education levels.  Consequently, many informal settlement dwellers are involved in criminal behaviour or run kiosks that sell alcohol, cigarettes or energy-dense foods at lower prices. This exposes other inhabitants to unhealthy lifestyles that increase their risk of becoming alcohol-and tobacco-dependent, and overweight/obese, which are well-known leading risk factors of NCDs. 
The study population comprised adults (females and males) living in Diepsloot in Blocks I, J, K and L of Extension 2, who were aged [greater than or equal to] 20 years.
The following inclusion criteria were used in this study: (i) informal residents living in Diepsloot; (i) informal residents residing in either Block I, J, K, or L of Extension 2 in Diepsloot; and (iii) informal Diepsloot residents with a body mass index (BMI) [greater than or equal to] 18.5 kg/[m.sup.2].
Sampling and sample size
The primary study utilised a systematic random sampling method. Every 7th informal settlement resident who underwent BMI assessment at the health campaign site was invited to participate in the survey. The selection procedure started from the 4th participant (starting point randomly chosen between numbers 1 and 7). A total of 2 135 participants were surveyed by CARe.
As the dataset had many survey respondents (2 135), who also met the inclusion criteria of the study, and considering that the study did not require collecting primary data, all participants were included for analysis.
The following variables were included in this study: gender, age, marital status, race, nationality, ethnicity, employment status, dwelling type, number of people in household, educational level, and BMI. Perceived susceptibility (perceptions about obesity, family medical history, smoking, unhealthy eating and physical inactivity as risk factors that could increase their risk for NCDs) was assessed by five items, measured on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). Perceived severity (perceptions about the seriousness of NCDs, and its consequences on the domains of life) was assessed by five items measured on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). Perceived benefits (perceptions about the effectiveness of taking health-promoting actions) were assessed by seven items measured on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). Perceived barriers (perceptions about changing one's behaviours to prevent NCD risk) were assessed by seven items measured on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). Perceived cues to action (participants' engagement in healthy behaviours to improve their health) were assessed by six items measured on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). Perceived self-efficacy (perceived ability to understand risks of NCDs and its prevention, as well as ability to change behaviour) was measured by five items on a 5-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree).
The independent variable was age group. Participants were categorised in three mutually exclusive age groups (18-35 (young adults), 36-50 (middle-aged adults), and >51 (older adults) years) set a priori in accordance with the WHO age category guidelines for the sub-Saharan setting.  The dependent variable comprised risk perceptions of NCDs based on HBM constructs (perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy). We collapsed the 5-point Likert scale responses into binary measures. Strongly agree and agree responses were pooled together into agree, while strongly disagree, disagree and neutral responses were pooled together into disagree.
Validity and reliability
In the primary survey, two experienced reviewers evaluated the questionnaire for content and face validity. They assessed the grammar, syntax, appropriateness, relevance and flow of the questions. The reviewers confirmed that items within each HBM domain were suitable indicators of the constructs.
For this study, Cronbach's alpha test was performed to: (i) assess internal consistency of items within each construct; (ii) assess if questions within each HBM domain in the primary survey questionnaire all reliably measured the same latent variable; and (iii) identify and consider whether any of the items (specific questions within each construct), which lower the internal consistency of the scale, were to be removed from the analysis. Cronbach's alpha coefficient for all constructs ranged between 0.78 and 0.89 for all constructs (Table 1). Cronbach's alpha coefficient for all the constructs met the standard of acceptable level of reliability (coefficient value of [greater than or equal to] 0.70) as recommended in the literature. [31,32]
The given dataset was in the format of an Excel spreadsheet, which we exported to STATA version 13 (Stata Corp., USA) to check for data completeness, outliers' missing data, grouping and coding variables and performing statistical analysis accordingly. All data were stratified by age groups. The data were found to be complete, with no missing variables. We collapsed variables with 5-point Likert scale responses into binary measures. Strongly agree and agree responses were grouped into agree, while strongly disagree, disagree and neutral responses were grouped into disagree. These were coded as follows: 0 = agree and 1 = disagree.
Descriptive statistics were used to summarise and describe the data. Numerical data were summarised as means and standard deviations (SDs). Categorical data were summarised as frequencies and percentages. Pearson's [chi square] test was used to examine whether differences in baseline characteristics between young adults, middle-aged adults and older-aged adults were statistically significant. Pearson's [chi square] test of independence was used to examine the association between dependent and independent variables. A p-value <0.05 was considered statistically significant.
Ethical clearance for this study was sought from Monash University Human Research Ethics Committee (ref. no. CF14/2803-2014001558) as part of a larger study. Authorisation to use the dataset for this study was given by CARe. The data were used anonymously.
Baseline characteristics of study participants
Table 2 shows the baseline characteristics of survey participants stratified by age groups. A total of 2 135 participants were included in the analysis, of whom 1 154 (54.05%) were females. Most of the survey participants were 18-35 years old (n=1 526, 71.48%) and the mean age was 31.2 (SD 9.87) years. The majority of the participants were from black communities (n=2 111, 98.88%) with poor socioeconomic status, living in informal houses (shacks) (n=1 740, 81.50%). Gender (p<0.001), employment status (p=0.025), nationality (p=0.001), marital status (p<0.001), type of dwelling (p=0.001), highest education qualification (p<0.001) and monthly income (p=0.001) had a statistically significant variation between young, middle-aged and older-aged adult participants (Table 2).
Association between age group and HBM constructs
This section reports analysis of the [chi square] test of independence comparing association between risk perceptions of NCDs and age group as a factor.
In the perceived susceptibility construct, variables were independently associated with age group as a factor. These variables include family medical history ([chi square] (2)=6.2020, p=0.045) and smoking ([chi square] (2)=11.0176, p=0.004). These factors were significantly associated with age groups, with more middle-aged adults perceiving these as risk factors that would increase their risk of NCDs than young and older-aged adults.
In the perceived severity construct, variables were independently associated with age group as a factor. These variables impact on life and family ([chi square] (2)=10.9161, p=0.004) and adverse thoughts ([chi square] (2)=7.5548, p =0.023). They were significantly associated with age group, with more middle-aged adults perceiving these factors as risks of NCD morbidities than young and older-aged adults.
In the perceived benefits construct, six variables were independently associated with age group as a factor. These variables include physical activity ([chi square] (2)=11.6180, p=0.003), healthy lifestyle ([chi square] (2)=20.0129, p<0.001), weight management ([chi square] (2)=28.4892, p<0.001), not smoking ([chi square] (2)=7.1273, p=0.028), and regular health check-ups ([chi square] (2)=24.0826, p<0.001). These factors were significantly associated with age group, with more older-aged adults perceiving the usefulness of these factors as NCD preventive measures compared with young and middle-aged adults.
The perception of health check-ups as a time-consuming process was significantly associated with age group ([chi square] (2)=13.5714, p=0.001). This association was greater among young adults compared with middle-aged and older-aged adults.
Perceived cues to action
The perception of eating small portions to improve health was significantly associated with age group ([chi square] (2)=13.7513, p=0.001). This association was greater among middle-aged adults compared with young and older-aged adults.
The perception of having confidence to understand NCD risks and prevention was significantly associated with age group ([chi square] (2)=17.3554, p<0.001). This association was higher among middle-aged adults compared with young and older-aged adults. There was a statistically significant association between age group and the perception of having health information to understand risks of NCDs and its prevention ([chi square] (2)=6.3745, p=0.041). The association was greater among middle-aged adults than young and older-aged adults. The perception of having the ability to actively work on health to prevent risk of NCDs was significantly associated with age group ([chi square] (2)=17.3554, p<0.001). This association was higher among middle-aged adults than young and older-aged adults.
Distribution of risk perceptions of NCDs stratified by age groups
Table 3 compares the distribution of risk perceptions of NCDs by age groups.
Significant variations between young adults, middle-aged adults and older-aged adults were found in the perceived susceptibility construct. The proportion of perceived susceptibility to risks of NCDs among middle-aged adults was greater than for young and older-aged adults, and the difference was statistically significant. A greater proportion of middle-aged adults compared with young and older-aged adults perceived family history (74.00% v. 72.74% v. 62.39%, p=0.045) and smoking (83.80% v. 77.20% v. 74.31%, p=0.004) as risk factors that would increase their risk of NCDs. The differences were statistically significant.
A higher proportion of older-aged adults than young and middle-aged adults perceived effects on life and family (89.91% v. 77.39% v. 75.40%, p=0.004) and adverse thoughts (89.91% v. 82.50% v. 79.20%, p=0.023) as risks of NCD morbidities. The differences were statistically significant. Significantly more middle-aged adults compared with young and older-aged adults perceived change in outlook as a risk of NCD morbidities (82.40% v. 74.71% v. 70.64%, p=0.001). The differences were statistically significant (Table 3).
The perceived benefits of adopting health-promoting behaviours with regard to the risk of NCDs were greater in middle-aged adults than young and older-aged adults. The differences were statistically significant. More middle-aged adults than young and older-aged adults perceived the usefulness of physical activities (84.40% v. 82.31% v. 70.64%, p=0.003), healthy lifestyle (87.20% v. 82.90% v. 69.72%, p<0.001), weight management (89.00% v. 83.09% v. 68.81%, p<0.001), not smoking (84.60% v. 81.06% v. 74.31%, p=0.028), and regular health check-ups (87.20% v. 80.01% v. 68.81%, p<0.001) as effective NCD prevent ive measures. The differences between age groups in each of these variables were statistically significant (Table 3).
Among the potential barriers, more young adults compared with middle-aged and older-aged adults perceived health check-ups as a time-consuming process to prevent risks of NCDs (59.31% v. 58.00% v. 41.28%, p=0.001). The difference was statistically significant. This barrier prevents them from adopting NCD preventive measures and understanding that health check-ups have an influence on NCDs (Table 3).
Perceived cues to action
The perception of eating small portions to improve health was greater among middle-aged adults than young and older-aged adults (40.00% v. 74.97% v. 61.47%, p=0.001). The difference was statistically significant.
The proportion of perceived self-efficacy to the risks of NCDs among middle-aged adults was higher than in young and older-aged adults. The differences were statistically significant. There were more middle-aged adults than young and older-aged adults who perceived to have the confidence (78.40% v. 74.97% v. 61.47%, p<0.001) and health information (84.00% v. 79.10% v. 77.06%, p=0.041) to understand the risks of NCDs and its prevention, as well as the ability to actively work on a healthy lifestyle (80.40% v. 77.85% v. 55.96%, p<0.001). The differences were statistically significant (Table 3).
The purpose of this study was to examine age-group differences in risk perceptions of NCDs among adults in Diepsloot township. Studies assessing this topic in African settings are scarce, with none in SA. Therefore, studies conducted in other countries were used to support the results of this study. Although several studies across developed and developing countries have examined knowledge, attitude and practice of preventive measures among individuals at increased risk of different NCDs, particularly breast cancer in females, [10,11,18,32] diabetes, [14,21] and CVDs, [12,13,16,33-38] this is the first identifiable study that explored the relationship between age groups and risk perceptions of NCDs, based on all the constructs of the HBM.
Knowledge is factual, objective and justified truth. Beliefs, however, are mental representations of one's thoughts from a cognitive point of view towards the likelihood of something being true.  Considering that: (i) biochemistry of knowledge and beliefs are different; and (ii) to our knowledge no literature has focused on individuals of specific age groups and their health beliefs of risk perceptions of NCDs, studies exploring similar concepts but in relation to NCD knowledge and attitude in particular cannot be used to support the results of this study. Instead, the results are explained by the HBM and other related theoretical frameworks. The results are unique findings that in many ways may influence existing and new NCD policies and programmes in SA.
Our study found that more young adults perceived the risks of NCDs to be lower than middle-aged and older-aged adults --predominantly family history and smoking as risk factors of NCDs, negative impact on life and family, outlook, and adverse thoughts as effects of NCDs morbidities. This could be because they lacked more confidence with regard to health information to understand the risks of NCDs and their prevention than middle-aged and older-aged adult participants. Furthermore, more young adults did not have a matriculation qualification than middle-aged and older-aged adults. This is concerning, as a low level of literacy is one of the leading factors influencing unhealthy and risky behaviours.  As unemployment and alcohol and drug abuse are some of the most pressing social problems among youths in Diepsloot,  it is possible that young adults are least exposed to NCD information as they go out to look for money to be able to purchase alcohol and drugs. Hence, they are less likely to access services such as HIV testing and counselling, which also provide NCD-related knowledge.
Poor perception of smoking as a risk factor of NCDs and the usefulness of not smoking as an effective NCD preventive measure among young adults is of concern. In accordance with the theory of invincibility, young adults often believe themselves to be invincible, which leads to the perception that somehow the consequences of their engagement in a particular behaviour will not happen, as they are invulnerable.  Therefore, they often underestimate the risks associated with their behaviour (i.e. smoking). There is therefore a need to consider revising SA's existing tobacco policy by integrating outreach and public health education programmes involving community leaders to reach out to more young adults in informal settlements. These are cost-effective and successful methods of escalating knowledge and awareness, [42,43] chronic obstructive pulmonary disease and lung cancer. It is noteworthy that the likelihood of perceiving oneself at greater risk escalates when the presence of a risk factor is known.  Risk indicators, for instance the number of cigarettes smoked, account for 24.0% variation in perceived risk.  However, the primary survey questionnaire did not capture the participants' smoking behaviours. Therefore, there is a possibility that many young adults may not have been smokers and did not perceive smoking as a risk of NCDs.
Regular health check-ups are known to be effective in detecting and treating chronic diseases at an early stage. We found that young adults were less likely to perceive the usefulness of regular health assessments as an effective NCD preventive measure than middleaged and older-aged adults. This could be because young adults were more likely to perceive health assessments as a time-consuming process compared with middle-aged and older-aged adults, which indicates a need to increase efforts to reach out to more young adults and educate them about the benefits associated with routine health check-ups.
Young adults were less likely to eat small-portion meals to improve their health compared with middle-aged and older-aged adults. This could be because they were less likely to perceive the usefulness of weight management in preventing NCDs, which indicates that poor risk perceptions of NCDs have a negative impact on adoption of NCD preventive measures. There is a possibility that this observed difference could have been confounded by gender, as perceptions of weight management varies by gender in African culture, where being overweight/obese has positive connotations. [45,46] According to an African-based study in Khayelitsha township in Cape Town, SA, Puoane et al.  revealed that being overweight was more ideal and preferred among black women because it signified that one is being taken good care of by one's spouse. Alternatively, from a male's point of view, it inferred that such women are healthier, capable to stir large pots, and would not be blown away by the strong Cape winds; desirability by males escalates as females become obese.  In light of this, our study did not account for gender.
Study strengths and limitations
This was the first study describing specific age-group differences in risk perceptions of NCDs in an informal settlement in SA. There had been a case study in Diepsloot describing the risk perceptions of NCDs using the HBM, but age-group differences were not reported.  However, our study has several limitations. It had a cross-sectional design, which did not allow making any causal inferences on the established associations found. The p-values only indicate existence of a relationship between two variables--not the strength of the relationship or which relationship is the most significant in each HBM construct.  The method of selecting the study participants in the primary survey might have introduced selection bias, assuming that those not selected because they did not undergo BMI assessment might have reported risk perceptions of NCDs differently. Moreover, data were self-reported; therefore, some survey participants might have over-reported socially desirable perceptions and behaviours.
More young adults had poor risk perceptions of NCDs than middle-aged and older-aged adults in Diepsloot township. They were less likely to adopt healthy behaviours compared with middle-aged and older-aged adults, which indicates that poor risk perceptions can negatively impact the uptake of prevention measures. This could be due to the misconception of the concept of invulnerability, possibly resulting from limited exposure to NCD-related information among young adults to understand NCD-related risks and prevention. This highlights the need to increase public health education programmes to increase outreach, NCDs knowledge and uptake of prevention measures, particularly among young adults. Further research could investigate gender differences in risk perceptions of NCDs among this population, which could provide a better understanding or possible explanation of the less positive trends in decrease of NCD-related mortalities among the younger population.
Acknowledgements. We would like to thank the CARe organisation, Johannesburg, SA, for providing the dataset used in this study. Author contributions. ZK: conceptualised the study, performed the statistical analysis of the secondary data, interpreted the data and results, alongside writing and revising the manuscript and scientific content. NK: supervised the study, provided essential constructs for the writing of the manuscript and interpretation of the results, and supervised the revision of important scientific content. NT: gave advice on the presentation of the results and editing of the manuscript text. All authors read and approved the final version of the manuscript.
Conflicts of interest. None.
[1.] Richards NC, Gouda HN, Durham J, Rampatige R, Rodney A, Whittaker M. Disability, noncommunicable disease and health information. Bull World Health Organ 2016;94(3):230-232. https://doi.org/10.2471/BLT.15.156869
[2.] World Health Organization. Global Status Report on Noncommunicable Diseases 2014. Geneva: WHO, 2014:1-176.
[3.] World Health Organization. Global Status Report on Noncommunicable Diseases 2010. Italy: WHO, 2011:1-176.
[4.] Hunter D, Reddy SK. Noncommunicable diseases. N Engl J Med 2013;369(14):1336-1343. https://doi. org/10.1056/NEJMra1109345
[5.] Beaglehole R, Bonita R, Horton R, et al. Priority actions for the non-communicable disease crisis. Lancet 2011;377(9775):1438-1447. https://doi.org/10.1016/S0140-6736(11)60393-0
[6.] World Health Organization. NCD Profile South Africa. Geneva: WHO, 2014:1-210.
[7.] Hofman K. Non-communicable diseases in South Africa: A challenge to economic development. S Afr Med J 2014;104(10):647. https://doi.org/10.7196/samj.8727
[8.] Australian Institute of Health and Welfare. Trends in Coronary Heart Disesase Mortality: Age Groups and Populations: Cardiovascular Disease Series. Canberra: AIHW, 2014:1-38.
[9.] Onoruoiza SI, Musa A, Umar BD, Kunle YS. Using health belief model as an intervention to noncompliance with hypertension information among hypertensive patient. Int Organ Sci Res J Human Soc Sci 2015;20(9):11-16. https://doi.org/10.9790/0837-20951116
[10.] Abolfotouh MA, BaniMustafa AA, Mahfouz AA, Al-Assiri MH, Al-Juhani AF, Alaskar AS. Using the health belief model to predict breast self examination among Saudi women. BioMed Central Public Health 2015;15(1):1-12. https://doi.org/10.1186/s12889-015-2510-y
[11.] Dundar PE, Ozmen D, Ozturk B, et al. The knowledge and attitudes of breast self-examination and mammography in a group of women in a rural area in western Turkey. BioMed Central Cancer 2006;6(43):1-9. https://doi.org/10.1186/1471-2407-6-43
[12.] Vaidya A, Aryal UR, Krettek A. Cardiovascular health knowledge, attitude and practice/behaviour in an urbanising community of Nepal: A population-base d cross-sectional study from Jhaukhel-Duwakot Health Demographic Surveillance Site. BioMed Central J Open 2013;3(10):e002976. https:// doi.org/10.1136/bmjopen-2013-002976
[13.] Khan MS, Jafary FH, Jafar TH, et al. Knowledge of modifiable risk factors of heart disease among patients with acute myocardial infarction in Karachi, Pakistan: A cross sectional study. BioMed Central Cardiovasc Dis 2006;6(18):1-9. https://doi.org/10.1186/1471-2261-6-18
[14.] Aljoudi AS, Taha AZ. Knowledge of diabetes risk factors and preventive measures among attendees of a primary care center in eastern Saudi Arabia. Ann Saudi Med 2011; 1(29): 15-19. https://doi. org/10.4103/0256-4947.51813
[15.] Song Y, Ma W, Yi X, et al. Chronic diseases knowledge and related factors among the elderly in Jinan, China. PLoS ONE 2013;8(6):e68599. https://doi.org/10.1371/journal.pone.0068599
[16.] Mochari-Greenberger H, Miller KL, Mosca L. Racial/ethnic and age differences in women's awareness of heart disease. J Women's Health 2012;21(5):476-480. https://doi.org/10.1089/jwh.2011.3428
[17.] Winham DM, Jones KM. Knowledge of young African American adults about heart disease: A cross-sectional survey. BioMed Central Public Health 2011;11(248):1-11. https://doi.org/10.1186/1471-2458-11-248
[18.] Agboola AOJ, Deji-Agboola AM, Oritogun KS, Musa AA, Oyebadejo TY, Ayoade BA. Knowledge, attitude and practice of breast self examination in female health workers in Olabisi Onabanjo University Teaching Hospital, Sagamu, Nigeria. Int Med J Malaysia 2009;8(1):5-10.
[19.] Mosca L, Benjamin EJ, Berra K, et al. Effectiveness-based guidelines for the prevention of cardiovascular disease in women--2011 update. J Am Coll Cardiol 2011;57(12):1404-1423. https://doi.org/10.1016/j. jacc.2011.02.005
[20.] Tshuma N. Perceptions of risk of non-communicable diseases associated with overweight and obesity among individuals living in Diepsloot informal settlements, Johannesburg, South Africa. S Am J Public Health 2014;2(4):629-665.
[21.] Dorman JS, Valdez R, Liu T, et al. Health beliefs among individuals at increased familial risk for type 2 diabetes: Implications for prevention. Diabet Res Clin Pract 2012;96(2):156-162. https://doi. org/10.1016/j.diabres.2011.12.017
[22.] World Health Organization. Global Action Plan for the Prevention and Control of Noncommunicable Diseases 2013-2020. Geneva: WHO, 2013:1-102.
[23.] Islam SMS, Purnat TD, Phuong NTA, Mwingira U, Schacht K, Froschl G. Non-communicable diseases (NCDs) in developing countries: A symposium report. Global Health 2014;10(81):1-7. https://doi. org/10.1186/s12992-014-0081-9
[24.] Abegunde DO, Mathers CD, Adam T, Ortegon M, Strong K. The burden and costs of chronic diseases in low-income and middle-income countries. Lancet 2007;370(9603):1929-1938. https://doi. org/10.1016/S0140-6736(07)61696-1
[25.] Levitt NS, Peer N, Steyn K, et al Increased risk of dysglycaemia in South Africans with HIV; especially those on protease inhibitors. Diabet Res Clin Prac 2016;119:41-47. https://doi.org/10.1016/j. diabres.2016.03.012
[26.] Oni T, Youngblood E, Boulle A, McGrath N, Wilkinson R, Levitt NS. Patterns of HIV, TB, and noncommunicable disease multi-morbidity in peri urban South Africa--a cross sectional study. BioMed Central Infect Dis 2015;15(20):1-8. https://doi.org/10.1186/s12879-015-0750-1
[27.] Collins DL, Leibbrandt M. The financial impact of HIV/AIDS on poor households in South Africa. AIDS 2007;21(suppl 7):S75-S81. https://doi.org/10.1097/01.aids.0000300538.28096.1c
[28.] Johannesburg Development Agency. Diepsloot Ready for Development. Johannesburg: JDA, 2013:1-8.
[29.] Statistics South Africa. Census 2011 Fact sheet. Pretoria: Statistics South Africa, 2011:1-2.
[30.] World Health Organization. Health Statistics and Health Information Systems: Definition of an Older or Elderly Person. Proposed Working Definition of an Older Person in Africa for the MDS Project. Geneva: WHO, 2001.
[31.] DeVon HA, Block ME, Moyle-Wright P, et al. A psychometric toolbox for testing validity and reliability. J Nursing Scholar 2007;39(2):155-164. https://doi.org/10.1111/j.1547-5069.2007.00161.x
[32.] Wong RK, Wong ML, Chan YH, Feng Z, Wai CT, Yeoh KG. Gender differences in predictors of colorectal cancer screening uptake: A national cross sectional study based on the health belief model. BioMed Central Public Health 2013;13(677):1-12. htttps://doi.org/10.1111/j.1547-5069.2007.00161.x
[33.] Abbaszadeh A, Borhani F, Asadi N. Effects of health belief model-based video training about risk factors on knowledge and attitude of myocardial infarction patients after discharge. J Res Med Sci 2011;16(2):195-200.
[34.] Ali NS. Prediction of coronary heart disease preventive behaviors in women: A test of the Health Belief Model. Women Health 2002;242(2002):83-96. https://doi.org/10.1300/J013v35n01_06
[35.] Oli N, Vaidya A, Subedi M, Krettek A. Experiences and perceptions about cause and prevention of cardiovascular disease among people with cardiometabolic conditions: Findings of in-depth interviews from a peri-urban Nepalese community. Global Health Action 2014;7(1):1-11. https://doi.org/10.3402/ gha.v7.24023
[36.] Katz DA, Graber M, Birrer E, et al Health beliefs toward cardiovascular risk reduction in patients admitted to chest pain observation units. Acad Emerg Med 2009;16(5):379-387. https://doi.org/10.1111/ j.1553-2712.2009.00383.x
[37.] Tovar E, Clark MC. Knowledge and health beliefs related to heart disease risk among adults with type 2 diabetes. J Am Assoc Nurse Pract 2014;27(6):1-7. https://doi.org/10.1002/2327-6924.12172
[38.] Choi S, Rankin S, Stewart A, Oka R. Perceptions of coronary heart disease risk in Korean immigrants with type 2 diabetes. Diabet Educ 2013;34(3):484-492. https://doi.org/10.1177/0145721708316949
[39.] Rao TSS, Asha MR, Rao KSJ, Vasudevaraju P. The biochemistry of belief. Ind J Psychiat 2009;51(4):239-241. https://doi.org/10.4103/0019-5545.58285
[40.] Hosseinpoor AR, Bergen N, Kunst A, et al Socioeconomic inequalities in risk factors for non communicable diseases in low-income and middle-income countries: Results from the World Health Survey. BioMed Central Public Health 2012;12(1):912. https://doi.org/10.1186/1471-2458-12-912
[41.] Killgore WDS, Kelley A, Balkin TJ. So you think you're bulletproof: Development and validation of the Invincibility Belief Index (IBI). Mil Med 2010;175(7):499-508. https://doi.org/10.7205/MILMED-D-09-00240
[42.] Schroetter SA, Peck SD. Women's risk of heart disease: Promoting awareness and prevention--a primary care approach. Medsurg Nursing (Lond) 2008;17(2):107-114.
[43.] Tavassoli E, Reisi M, Javadzad SH, et al. The effect of education on the improvement of fruits and vegetables consumption aiming to preventing colorectal cancer. Med J Islam Rep Iran 2015;29(183):1-7.
[44.] Chen W, Yu Y, Glaser K. The knowledge and attitudes of coronary heart disease prevention among middle and older aged people in a community in Taipei. Lay people's knowledge and attitudes of CHD prevention. Taiwan Geriat Gerontol 2009;4(4):251-262.
[45.] Devanathan R, Esterhuizen TM, Govender RD. Overweight and obesity amongst Black women in Durban, KwaZulu-Natal: A 'disease' of perception in an area of high HIV prevalence. Afr J Prim Health Care Fam Med 2013;5(1):1-7. https://doi.org/10.4102/phcfm.v5i1.450
[46.] Puoane T, Fourie JM, Shapiro M, Rosling L, Tshaka N. Obesity 'Big is beautiful'--an exploration with urban black community health workers in a South African township. S Afr J Clin Nutr 2005;18(1):6-15.
[47.] Greenland S, Senn S, Rothman KJ, et al Statistical tests, p-values, confidence intervals, and power: A guide to misinterpretations. Eur J Epidemiol 2016;31(4):337-350. https://doi.org/10.1007/s10654016-0149-3
Accepted 26 April 2017.
Z Kaba, (1) BPH; N Khamisa, (1) PhD; N Tshuma, (1,2) BSc, MBA
(1) Department of Public Health, School of Health Sciences, Monash South Africa, Johannesburg, South Africa
(2) Community AIDS Response, Johannesburg, South Africa
Corresponding author: Z Kaba (firstname.lastname@example.org)
Table 1. Internal consistency of the scales for HBM and risk perceptions of NCDs Items in the Alpha- HBM constructs scale, n coefficient Perceived susceptibility 5 0.8133 Perceived severity 5 0.8357 Perceived benefits 7 0.7791 Perceived barriers 7 0.8655 Perceived cues to action 5 0.7832 Perceived self-efficacy 6 0.8414 NCDs = non-communicable diseases; HBM = Health Belief Model. Table 2. Baseline demographic characteristic of study participants stratified by age groups Variables Total (N=2 135) Young adults (18-35 years, n=1 526) (71.48%) Mean age (SD) 32.1 (9.87) Gender, n (%) Female 1 154 (54.05) 867 (56.82) Male 981 (45.95) 659 (43.18) Ethnicity, n (%) Black 2 111 (98.88) 1 505 (98.62) Coloured 24 (1.12) 21 (1.38) Employment status, n (%) Employed 273 (12.79) 184 (12.06) Unemployed 1 862 (87.21) 1 342 (87.94) Nationality, n (%) South African 1 860 (87.12) 1 304 (85.45) Non-South African 275 (12.88) 222 (14.55) Marital status, n (%) Cohabitating 6 (0.28) 0 (0.00) Divorced 25 (1.17) 8 (0.52) Married 510 (23.89) 268 (17.56) Single 1 586 (74.29) 1 250 (81.91) Widow 8 (0.37) 0 (0.00) Type of dwelling, n (%) House 395 (18.50) 253 (16.58) Shack 1 740 (81.50) 1 273 (83.42) People in household, n (%) Alone 79 (3.70) 56 (3.67) 2 - 5 1 128 (52.83) 794 (52.03) 6 - 10 928 (43.47) 676 (44.30) Highest education qualification, n (%) Certificate 75 (3.51) 56 (3.67) Degree 37 (1.73) 31 (2.03) Diploma 77 (3.61) 54 (3.54) Matric 744 (34.85) 518 (33.94) No matric 1 202 (56.30) 867 (56.82) Monthly income, n (%) <ZAR1 000 1 862 (87.21) 1 342 (87.94) ZAR1 000 - 4 999 225 (10.54) 161 (10.55) > ZAR4 999 48 (2.25) 23 (1.51) BMI (%) Overweight/obese 1 285 (61.37) 923 (61.70) Normal weight 809 (38.63) 573 (38.30) Variables Middle-aged adults Older-aged adults (36 - 50 years, ([greater than or n=500) (23.42%) equal to] 51 years, n=109) (5.11%) Mean age (SD) Gender, n (%) Female 243 (48.60) 44 (40.37) Male 257 (51.40) 65 (59.63) Ethnicity, n (%) Black 497 (99.40) 109 (100.00) Coloured 3 (0.60) 0 (0.00) Employment status, n (%) Employed 80 (16.00) 9 (8.26) Unemployed 420 (84.00) 100 (91.74) Nationality, n (%) South African 455 (91.00) 101 (92.66) Non-South African 45 (9.00) 8 (7.34) Marital status, n (%) Cohabitating 6 (1.20) 0 (0.00) Divorced 8 (1.60) 9 (8.26) Married 200 (40.00) 42 (38.53) Single 281 (56.20) 55 (50.46) Widow 5 (1.00) 3 (2.75) Type of dwelling, n (%) House 121 (24.20) 21 (19.27) Shack 379 (75.80) 88 (80.73) People in household, n (%) Alone 23 (4.60) 0 (0.00) 2 - 5 280 (56.00) 54 (49.54) 6 - 10 197 (39.40) 55 (50.46) Highest education qualification, n (%) Certificate 19 (3.80) 0 (0.00) Degree 6 (1.20) 0 (0.00) Diploma 23 (4.60) 0 (0.00) Matric 212 (42.40) 14 (12.84) No matric 240 (48.00) 95 (87.16) Monthly income, n (%) <ZAR1 000 420 (84.00) 100 (91.74) ZAR1 000 - 4 999 58 (11.60) 6 (5.50) > ZAR4 999 22 (4.40) 3 (2.75) BMI (%) Overweight/obese 302 (61.13) 60 (57.69) Normal weight 192 (38.87) 44 (42.31) Variables p-value Mean age (SD) Gender, n (%) Female 0.000 * Male Ethnicity, n (%) Black 0.188 Coloured Employment status, n (%) Employed 0.025 * Unemployed Nationality, n (%) South African 0.001 * Non-South African Marital status, n (%) Cohabitating 0.000 * Divorced Married Single Widow Type of dwelling, n (%) House 0.001 * Shack People in household, n (%) Alone 0.043 2 - 5 6 - 10 Highest education qualification, n (%) Certificate 0.000 * Degree Diploma Matric No matric Monthly income, n (%) <ZAR1 000 0.001 * ZAR1 000 - 4 999 > ZAR4 999 BMI (%) Overweight/obese 0.714 Normal weight SD = standard deviation; BMI = body mass index. * Significant at p<0.05. Table 3. Distribution of risk perceptions of NCDs stratified by age groups Variables Young adults Middle-aged adults (18 - 35 years, (36 - 50 years, n=1 526) (72.48%), n=500) (23.42%), agree agree Perceived susceptibility, n (%) Being obese/overweight 1 169 (76.61) 377 (75.40) will lead me to getting NCDs My family medical 1 110 (72.74) 370 (74.00) history makes it likely to get NCDs Smoking makes it 1 178 (77.20) 419 (83.80) likely to get NCDs Unhealthy eating 1 149 (75.29) 389 (77.80) habits can make me get NCDs Physical inactivity 1 203 (78.83) 391 (78.20) can make me get NCDs Perceived severity, n (%) Having an NCD will 1 181 (77.39) 377 (75.40) have major effects on my life and family Having an NCD will 1 184 (77.59) 372 (74.40) have major effects on my work and income Having an NCD 1 183 (77.52) 385 (77.00) will cripple me Having an NCD will 1 140 (74.71) 412 (82.40) change my outlook Thought of having NCDs 1 259 (82.50) 396 (79.20) scares me Perceived benefits, n (%) Not having an NCD 1 198 (78.51) 381 (76.20) is beneficial Physical activities 1 256 (82.31) 422 (84.40) prevent NCDs Healthy lifestyle 1 265 (82.90) 436 (87.20) prevents NCDs Managing weight 1 268 (83.09) 445 (89.00) prevents NCDs Not smoking 1 237 (81.06) 423 (84.60) prevents NCDs Regular health 1 221 (80.01) 436 (87.20) check-ups will detect NCDs early Regular health 1 207 (79.10) 363 (72.60) check-ups are beneficial Perceived barriers, n (%) Very little can be 912 (59.76) 311 (62.20) done to prevent NCDs No treatment will be 800 (52.42) 292 (58.40) effective in curing NCDs Healthy foods 971 (63.63) 316 (63.20) are expensive Health check-ups 869 (56.95) 277 (55.40) are expensive Preparing healthy foods 901 (59.04) 271 (54.20) is time-consuming Health check-ups are 905 (59.31) 290 (58.00) time-consuming It is embarrassing to 838 (54.91) 262 (52.40) go for health check-ups Perceived cues to action, n (%) I eat a 1 097 (71.89) 304 (60.80) well-balanced diet I always follow 1 134 (74.31) 359 (71.80) medical orders to benefit my health I frequently make 1 112 (72.87) 370 (74.00) efforts to improve my health I exercise regularly, 1 137 (74.51) 385 (77.00) at least 3 times per week I avoid fatty foods 1 135 (74.38) 372 (74.40) I eat small-portion 1 144 (74.97) 392 (78.40) meals Perceived self-efficacy, n (%) I am confident about 1 183 (77.52) 399 (79.80) how to prevent chronic diseases I am able to actively 1 188 (77.85) 402 (80.40) work on a healthy lifestyle to prevent NCDs I attend health 1 179 (77.26) 380 (76.00) assessments to prevent NCDs I have information on 1 207 (79.10) 420 (84.00) how to prevent NCDs There is a lot I can 1 097 (71.89) 304 (60.80) do to reduce my chances of getting an NCD-related illness Variables Older-aged adults p-value ([greater than or equal to] 51 years, n=109) (5.11%), agree Perceived susceptibility, n (%) Being obese/overweight 79 (72.48) 0.567 will lead me to getting NCDs My family medical 68 (62.39) 0.045 * history makes it likely to get NCDs Smoking makes it 81 (74.31) 0.004 * likely to get NCDs Unhealthy eating 81 (74.31) 0.488 habits can make me get NCDs Physical inactivity 81 (74.31) 0.534 can make me get NCDs Perceived severity, n (%) Having an NCD will 98 (89.91) 0.004 * have major effects on my life and family Having an NCD will 87 (79.82) 0.261 have major effects on my work and income Having an NCD 83 (76.15) 0.928 will cripple me Having an NCD will 77 (70.64) 0.001 * change my outlook Thought of having NCDs 98 (89.91) 0.023 * scares me Perceived benefits, n (%) Not having an NCD 83 (76.15) 0.508 is beneficial Physical activities 77 (70.64) 0.003 * prevent NCDs Healthy lifestyle 76 (69.72) <0.001 * prevents NCDs Managing weight 75 (68.81) <0.001 * prevents NCDs Not smoking 81 (74.31) 0.028 * prevents NCDs Regular health 75 (68.81) <0.001 * check-ups will detect NCDs early Regular health 81 (74.31) 0.008 * check-ups are beneficial Perceived barriers, n (%) Very little can be 79 (72.48) 0.026 * done to prevent NCDs No treatment will be 74 (67.89) 0.001 * effective in curing NCDs Healthy foods 65 (59.63) 0.703 are expensive Health check-ups 59 (54.13) 0.735 are expensive Preparing healthy foods 56 (51.38) 0.068 is time-consuming Health check-ups are 45 (41.28) 0.001 * time-consuming It is embarrassing to 48 (44.04) 0.069 go for health check-ups Perceived cues to action, n (%) I eat a 78 (71.56) <0.001 * well-balanced diet I always follow 81 (74.31) 0.536 medical orders to benefit my health I frequently make 81 (74.31) 0.854 efforts to improve my health I exercise regularly, 84 (77.06) 0.481 at least 3 times per week I avoid fatty foods 75 (68.81) 0.433 I eat small-portion 67 (61.47) 0.001 * meals Perceived self-efficacy, n (%) I am confident about 67 (61.47) <0.001 * how to prevent chronic diseases I am able to actively 61 (55.96) <0.001 * work on a healthy lifestyle to prevent NCDs I attend health 75 (68.81) 0.125 assessments to prevent NCDs I have information on 84 (77.06) 0.041 * how to prevent NCDs There is a lot I can 78 (71.56) <0.001 * do to reduce my chances of getting an NCD-related illness NCDs = non-communicable diseases. * Significant at p<0.05.
|Printer friendly Cite/link Email Feedback|
|Author:||Kaba, Z.; Khamisa, N.; Tshuma, N.|
|Publication:||South African Medical Journal|
|Date:||Sep 1, 2017|
|Previous Article:||New-onset diabetes after transplant: incidence, risk factors and outcome.|
|Next Article:||Reversing disease.|