Stroke awareness and associated factors among hypertension and diabetes mellitus patients attending Tororo General Hospital, Eastern Uganda: a cross-sectional study

¹Department of Family Medicine, School of Medicine, College of Health Sciences, Makerere University, Kampala, Uganda, ²Department of Internal Medicine, School of Medicine, College of Health Sciences, Makerere University, Kampala, Uganda

^&Corresponding author
Nicholas Higenyi, Department of Family Medicine, School of Medicine, College of Health Sciences, Makerere University, Kampala, Uganda

Introduction    

Stroke is a significant global health problem that contributes to morbidity, mortality, and disability in both high- and low-income countries [1]. Stroke is second to ischemic heart disease among the most common causes of cardiovascular deaths worldwide [2]. In Uganda, according to the World Health Organization (WHO)-NCD country profile published in 2018, cardiovascular disease accounted for nearly 10% of all deaths in the year 2016, of which stroke is a principal contributor [3].

One of the major reasons for the increase in stroke as a cause of death is patients' lack of knowledge of the risk factors involved and warning signs, hence the delay in seeking treatment [4]. Studies have shown that awareness of stroke among patients and their caregivers is associated with reduced delay in seeking care. In a study in Korea, awareness by the patient/bystander that the initial symptom was stroke-related was highly associated with early arrival at the hospital [5]. Stroke is often accompanied by disabilities that adversely affect the well-being of both the survivors and their care providers. Studies have shown that stroke is preceded by many years of modifiable risk factors. The majority of cases of stroke are attributable to hypertension [6]. Arterial baroreflex dysfunction, oxidative stress, inflammation, and alterations in cerebral blood flow are the mechanisms through which hypertension causes stroke [7,8].

Another important risk factor for stroke is diabetes mellitus. The risk of stroke among diabetic participants is two to four times that of stroke among participants with normal glucose levels [9]. The mechanisms through which diabetes mellitus leads to stroke may include: vascular endothelial dysfunction, capillary basement membrane thickening, systemic inflammation, and accelerated premature arteriosclerosis [10]. Also, timely interventions among patients with stroke have been shown to reduce the risk of complications like disabilities and death [11,12]. Better knowledge and awareness of stroke among the patients and their family members has been demonstrated to be associated with timely treatment-seeking behavior [13]. Therefore, improving awareness of stroke among the at-risk patients like patients with diabetes mellitus and hypertension, may not only reduce pre-hospital delay, improve treatment-seeking behavior, and increase the chance of getting urgent medical attention but also increase participation in preventive programs.

Whereas awareness of stroke appears to be low in Central Uganda [14], it is not known whether awareness and associated factors are any different among hypertensive and diabetic patients, as hypertension and diabetes are important risk factors for stroke, especially in Eastern Uganda. Studies done elsewhere in Africa have suggested an individual's age, type of place of residence, sex, and level of education as the major factors influencing public awareness of stroke [15]. Other studies also suggested that having diabetes mellitus, comorbidities like HIV, and having a history of stroke or caring for someone with stroke are key determinants of awareness of stroke [16]. There is a high burden of hypertension in Uganda, where 1 in every 4 adults is estimated to have high blood pressure [17]. Further, about 7.4% of adults between the ages of 35 and 60 years in Eastern Uganda have diabetes mellitus [18].

However, little is known about the level of knowledge and associated factors of stroke prevention, warning signs, and response among this population at risk. This information is essential for the design of health promotive and stroke preventive and response strategies. This study therefore sought to assess stroke awareness among hypertensive and diabetic patients attending Tororo General Hospital, Tororo Municipality in Eastern Uganda.

Methods     

Study design and setting: this cross-sectional study was conducted at Tororo General Hospital in the non-communicable disease clinic of the outpatient department. In total, the clinic has about 620 clients with hypertension and 120 clients with diabetes mellitus. The clinic serves patients from Tororo and the neighboring districts. Tororo District is located about 220 km East of Uganda's capital city, Kampala, and borders Western Kenya.

Study population and eligibility: the study population was adult patients diagnosed with hypertension and/or diabetes mellitus receiving care at Tororo General Hospital between May and December 2024 who consented to participate in the study. All adult patients diagnosed with hypertension and/or diabetes aged at least 18 years attending care at Tororo General Hospital were eligible for enrolment in the study. Clients with cognitive impairment were excluded from the study as they could neither provide informed consent nor respond appropriately to the questions in the questionnaire.

Sample size determination: the sample size for the study was determined using the Kramer-Greenhouse formula. The calculated sample size based on a similar study in Nigeria [19] was 640 participants. However, the clinic had about 740 registered clients. The sample size was adjusted for a finite population to 190 participants. After adjusting for an estimated noncompletion rate of 10%, the final calculated minimum sample size was 212 participants.

Data collection: we used systematic random sampling with an interval of 3 (derived from dividing the number of clients registered in the clinic (740) by the sample size (212)) to recruit participants as they exited the clinic on each day of the clinic. A total of 212 study participants were enrolled in the study. Data were collected using a stroke awareness questionnaire adapted from studies on stroke awareness in Ireland [4], Uganda [20] and Nigeria [19]. The questionnaire is reported to have good internal consistency. The first section consisted of participant socio-demographics: age, sex, marital status, average monthly income, highest level of education attained, tribe, and length of time since enrollment in the clinic. The second section measured respondents' knowledge of stroke by recognition of the organ involved in the disease, risk factors, warning signs, and information sources.

Study variables

Dependent variable: the primary outcome variable was awareness of stroke. It was a categorical variable measured by participants' knowledge of stroke warning signs.

Independent variables: the independent variables included the following: sociodemographic variables including age in completed years, sex of the respondents, marital status, highest level of education and average monthly income.

The individual variables included: participant's age, level of education, marital status, employment, monthly income, home distance from hospital, HIV status, source of stroke-related information, and length of time as a diabetes/hypertension client in the clinic. The health facility-related variables included: whether a client has ever received stroke-related health education at the facility, and whether they had ever seen/received any stroke-related information education and communication (IEC) materials from the facility.

Statistical analysis: the data were analyzed using Stata software version 16. Categorical variables were summarized using frequencies and proportions while continuous variables were summarized with means and corresponding standard deviations or medians with corresponding interquartile ranges. Before bivariable and multivariable analyses, the dependent variable - level of awareness of stroke was categorized into a binary variable with two groups. All those with knowledge of at least 2 stroke warning signs were categorized into one group of "good awareness of stroke" (this group was coded as 1). Respondents who had knowledge of 0 and 1 stroke warning sign were categorized together as having "poor awareness of stroke" (this was coded as 0). Chi-square test was used for bivariable analysis of factors associated with awareness of stroke. The factors associated with awareness of stroke at a cut-off p-value of 0.2 were assessed for collinearity. Non-collinear variables and factors considered important for the study were assessed using modified Poisson regression to determine the factors independently associated with awareness of stroke at a critical p-value of 0.05.

Ethical approval and consent to participate: ethical clearance to conduct the study was granted by Makerere University School of Medicine Research and Ethics Committee (approval number: SOMREC, 2023-823). Permission to conduct the study was also granted by the medical superintendent of Tororo General Hospital. Informed consent was obtained, and participation was fully based on the willingness of participants. The respondents were allowed to refuse or discontinue participation at any time. Information was recorded anonymously. Confidentiality and privacy were ensured throughout the study.

Results     

The majority of the participants were female (57%). The mean and median age were 53 (SD 12.8) years. More than 4 in every 5 (84.4%) participants were married. Nearly 65% of the participants either had attained only primary level education or no formal education at all. Nearly 7 out of ten participants reported being employed or doing something that earned them money. The median estimated monthly income of the study participants was Ug. shs 100000 (IQR215000). The median distance between the participants' residence and the hospital was 7.0 (IQR 29.0) kilometers. The median duration for which the participants had attended the NCD clinic was 3.0 (IQR 14.0) years. Nearly three-quarters of the participants were attending the clinic as clients with hypertension, while the rest were either diabetic or had been diagnosed with both diabetes mellitus and hypertension. About 10% of the participants were also living with HIV. The social and demographic characteristics of the study participants are shown in Table 1.

Knowledge of stroke among hypertensive and/or diabetes mellitus patients attending the Tororo General Hospital Non-communicable Disease Clinic: half of the study participants could not define what stroke is, and 9 out of every ten participants did not know what a minor stroke or transient ischemic attack is. Only about 30% of the study participants could list at least two risk factors for developing stroke. The majority of the participants (64.6%) could not name the organ of the body affected by stroke. Nearly 80% of the participants admitted that stroke is preventable. One hundred and seventy-six of the participants reported that stroke had an effect on daily activities.

Prevalence of stroke awareness among hypertensive and/or diabetes mellitus patients attending the Non-communicable Disease Clinic of Tororo General Hospital: only 94 out of the 212 participants could identify at least two symptoms or warning signs of stroke. The proportion of participants with awareness of stroke at Tororo General Hospital was 44.3% (95% CI 37.8% -51.1%).

Factors associated with stroke awareness among clients with hypertension and/or diabetes mellitus attending Tororo General Hospital Non-communicable Disease Clinic: the factors associated with stroke awareness after bivariable analysis at the 95% level of significance were: the participants' source of stroke-related information (p<0.01) and the length of time as clients in the NCD clinic (p<0.04). The bivariable analysis of factors associated with stroke awareness among the clients with hypertension and/or diabetes attending the TGH NCD clinic using chi-square or Fisher's exact test is shown in Table 2. Participants' level of education was not significantly associated with stroke awareness at the bivariable level (p<0.07). Awareness of stroke was not significantly associated with the participants' sex at the bivariable level (p<0.71). The distance of the participants' home from the clinic was not significantly associated with stroke awareness at the bivariable level (p<0.46). Participants' estimated monthly income (p<0.64) and age (p<0.11) were not significantly associated with stroke awareness at the bivariable level. Stroke awareness did not differ significantly among participants who reported having health education and those who reported not having any stroke-related health education (p<0.08). Stroke awareness did not vary significantly among patients diagnosed with hypertension or diabetes mellitus (p<0.053), and neither did it vary significantly by participants' HIV serostatus (p<0.29).

Eight variables from bivariable analysis were considered for multivariable analysis using a p-value threshold of ≤ 0.2, while the rest were conceptually important confounders. Modified Poisson regression was used for multivariable analysis as the log-binomial regression model failed to achieve convergence. Three factors were significantly associated with awareness of stroke: sources of stroke-related information (aPR=2.13 95% CI 1.27-3.59, p<0.00), length of time as NCD clinic client (aPR=2.05 95% CI 1.31-3.20, p<0.00) and whether the client had received stroke-related health education (aPR=0.56 95% CI 0.33-0.93, p<0.03). The level of stroke awareness among clients who had attended the clinic for at least 5 years was 1.7 to 2 times that of clients who had attended the clinic for less than two years. The proportion of stroke awareness among clients who had never received stroke-related health education was almost half that of clients who reported receiving stroke-related health education. Also, the proportion of stroke awareness among clients who reported radio and television as their sources of stroke-related information was 2 to 2.7 times that of stroke awareness among clients who reported that health care providers were their sources of stroke-related information. Sex, age, level of education, non-communicable disease diagnosis, and whether the participant had seen any stroke-related IEC materials were not associated with awareness of stroke. The multivariable analysis of factors associated with stroke awareness among clients with hypertension and/or diabetes mellitus attending the Tororo General Hospital Non-communicable Disease Clinic is shown in Table 3.

Discussion     

This study assessed the level of awareness of stroke and associated factors among diabetic and hypertensive patients in Eastern Uganda. Main findings are that the level of awareness of stroke is low. Client stroke-related health education, participants' source of stroke-related information, and length of time as an NCD clinic client were the factors independently associated with stroke awareness. This information is essential for future strategies and campaigns to promote awareness of stroke and design prevention strategies so as to reduce stroke-related morbidity and mortality.

This study found that only 44% of the participants had awareness of stroke. This was very low, as these are the patients who are at the highest risk of suffering from stroke. This implies that the majority of the patients would not recognize that they were having a stroke early enough so as to seek early treatment. The findings are similar to those of a study in rural and urban Uganda where only 47% of the participants knew at least two stroke warning signs and symptoms [21]. The study findings are much higher than those of a cross-sectional survey in Greater Kampala where 17.7% of the participants knew at least 3 stroke warning signs [14]. The difference may be due to the difference in the number of warning signs used, as the current study considered knowledge of two warning signs. Also, the current study being among hypertensive and diabetic patients as compared to the one in Greater Kampala, the difference in study populations may also account for the difference in level of awareness, as participants from the NCD clinic are more likely to have been informed about stroke within the environment of the clinic as compared to the general population survey participants. The findings are considerably lower than those of a study in Accra, Ghana, on community awareness of stroke, where 59% of the participants identified at least two stroke warning signs and symptoms [22]. The difference could be due to the differences in sociodemographic and other determinants of stroke awareness between the two study populations.

This study found that having stroke-related health education was independently associated with awareness of stroke. Participants who reported receiving stroke-related health education were twice as likely to have awareness of stroke as compared to participants who reportedly had not received any stroke-related health education. This study finding is similar to the findings of a study in India where participants who had a professional source of stroke-related information were more likely to have awareness of stroke [23]. This implies that health professionals should prioritize equipping patients with information about the possible complications of their various illnesses. This may not be overemphasized enough for diabetic and hypertensive patients, who are among the most at-risk population for developing stroke and its complications. The study findings are significantly different from those of a study in greater Kampala where there were no factors significantly associated with stroke awareness [14]. The difference in the findings may be due to differences in the study populations. The study in greater Kampala was conducted in the general population, while this study was among patients diagnosed with hypertension and/or diabetes mellitus who attend the NCD clinic of a general hospital.

In this study, participants who identified radio or television as their sources of stroke-related information were more likely to have awareness of stroke compared to participants who listed health workers, newspapers, or friends and relatives as their source of stroke-related information. This may be due to the fact that radio and television programs, when they address certain health topics, tend to provide detailed information to viewers and listeners and sometimes even permit listeners and viewers to call in and ask questions. It is, however, intriguing to find that such participants would be more informed about stroke than those who listed health workers as their source. It may point to the fact that health workers in their interaction with their clients have not paid adequate attention to providing detailed information about stroke and that the radio and television talk shows have done a better job. A study at two referral hospitals has suggested that knowledge of stroke may be poor among front-line health care providers as well [24]. However, radio and television talk shows are prone to biases in the information they give to their listeners and viewers, which tends to be influenced by the sponsors of the talk shows. The findings are similar to those of a study in South East Asia where sources of stroke-related information were significantly associated with knowledge of stroke and its mechanisms [25]. The finding is significantly different from the findings of studies in Greater Kampala and rural and urban Uganda where no factors were significantly associated with awareness of stroke [14,21]. The difference in the findings may be due to the differences in the study populations, with the studies in Greater Kampala and urban Uganda being population-based surveys as compared to this study, which was conducted among hypertensive and diabetic patients attending a general hospital non-communicable disease clinic.

Also, participants who had attended the non-communicable disease clinic for at least five years were more likely to have awareness of stroke as compared to clients who had attended the clinic for shorter periods. This may be due to the fact that the longer the client spent attending the clinic, the higher the chances that they would be exposed to the various sources of stroke-related information. Also, the clients who have attended the clinic longer tend to have higher chances of receiving stroke-related health education, which boosts their knowledge and awareness of stroke, as reported by studies in Ethiopia and Morocco [15,26].

In this study, age, sex, levels of education, and religion were not significantly associated with stroke awareness. This was in contrast to the findings in Accra, Ghana, where sociodemographic factors like age and religion were significantly associated with awareness of stroke [22]. It is also different from the findings of studies in Nigeria, Morocco, and Beirut, where level of education, sex and age were significantly associated with knowledge of stroke warning signs and symptoms [15,19,27].

Strengths of the study: according to published literature, this is among the rare studies assessing stroke awareness in Eastern Uganda, where Tororo District is situated. The staff of the clinic did not participate in interviewing the participants, which minimized the possibility of courtesy bias among the respondents, especially on items in the tool that involved giving information about the non-communicable disease clinic or the hospital. The use of an interviewer-administered questionnaire ensured the responses were meaningful as it enhances participants' understanding of the items in the tool by seeking clarification from the interviewers.

Limitations of the study: the study being health facility-based presents the first limitation, which has focused exclusively on people attending a public general hospital; they may be significantly different from hypertensive and diabetes mellitus patients who attend private clinics and hospitals in terms of awareness of stroke. Another limitation is the cross-sectional nature of the study, which measures only the current level of awareness of stroke and does not account for changes over time. The use of an interviewer-administered questionnaire may have resulted in courtesy bias on factors concerning the participants' care, like receiving health education and IEC materials. Also, most of the factors, like distance from the health facility, length of time as a client, and income, were self-reported with limited means of verification, which may have resulted in inaccurate measurements. The study defined awareness of stroke as knowledge of at least two warning signs of stroke, which limits comparability with similar studies in which awareness of stroke was defined as knowledge of at least 3 stroke warning signs. A lower threshold for awareness of stroke, however, facilitated multivariable analysis by reducing on the occurrence of cells with frequency counts of less than 5.

Conclusion     

Awareness of stroke among patients with diabetes and hypertension in Tororo is low. Use of stroke-related health education among patients with hypertension and diabetes, together with dissemination of stroke-related information through mass media like radio and television, as well as provider continuity of care, is likely to improve awareness and thereby promote early care seeking, hence reducing stroke-related morbidity and mortality. Non-communicable disease clinics ought to conduct routine stroke-related health education among patients with diabetes and/or hypertension to raise the level of awareness of stroke. The clinics ought to reach out to the public through mass media like local radio stations and television. The non-communicable disease clinics should prioritize retention of their clients in care, as this promotes sharing of knowledge and experiences about stroke from among the clients who may have been affected by stroke. The knowledge of health workers regarding stroke should be enhanced through routine continuous medical education about stroke and its complications, which they might in turn pass on to their clients during health education in the non-communicable disease clinics.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Nicholas Higenyi conceptualized the study, wrote the proposal, collected and analyzed the data, and wrote the manuscript; Jude Tadeo Onyango and Jane Frances Namatovu guided the process from conceptualization, proposal writing, data collection, and data analysis; Jude Tadeo Onyango and Mark Kaddu Mukasa supervised writing and edited the manuscript. All the authors read and approved the final version of this manuscript.

Acknowledgments     

This study was supported by the National Institutes of Health (NIH)/Makerere University College of Health Sciences Reducing Stroke Program. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NIH. The authors are grateful to the participants, research assistants, the staff and leadership of Tororo General Hospital for their time and assistance in data collection.

Tables     

Table 1: sociodemographic characteristics of study participants selected from hypertensive and/or diabetes mellitus patients attending the Non-communicable Diseases Clinic at Tororo General Hospital (Uganda) between May 2024 and December 2024 (N=212)

Table 2: bivariable analysis of factors associated with stroke awareness among study participants selected from hypertensive and/or diabetes mellitus patients attending Tororo General Hospital (Uganda) between May 2024 and December 2024 (N=212)

Table 3: multivariable analysis of factors associated with stroke awareness among study participants selected from hypertensive and/or diabetes mellitus patients attending the Non-communicable Diseases Clinic at Tororo General Hospital (Uganda) between May 2024 and Dec 2024, using modified Poisson regression (N=212)

References