AAS Open Res AAS Open Research 2515-9321 F1000 Research Limited London, UK 10.12688/aasopenres.12992.1 Research Article Articles Self-management and its associated factors among people living with diabetes in Blantyre, Malawi: a cross-sectional study

[version 1; peer review: 2 approved with reservations]

 Kwanjo Banda Chimwemwe Conceptualization Data Curation Formal Analysis Funding Acquisition Investigation Methodology Project Administration Resources Software Validation Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-9337-3583 a 1 2 Gombachika Belinda T. Conceptualization Methodology Supervision Validation Writing – Review & Editing 2 Nyirenda Moffat J. Conceptualization Investigation Methodology Supervision Validation Writing – Review & Editing 1 3 4 Muula Adamson Sinjani Conceptualization Methodology Resources Supervision Validation Writing – Review & Editing 1 4 5 School of Public Health and Family Medicine, University of Malawi College of Medicine, Blantyre, Malawi Medical and Surgical Nursing, University of Malawi, Kamuzu College of Nursing, Blantyre, Malawi Uganda MRC/UVRI Research Unit, Entebbe, Uganda NCD-Brite Consortium, University of Malawi College of Medicine, Blantyre, Malawi Africa Center of Excellence in Public Health and Herbal Medicine, University of Malawi College of Medicine, Blantyre, Malawi cbanda@cartafrica.org

No competing interests were disclosed.

 19 9 2019 2019 2 161 12 9 2019 Copyright: © 2019 Kwanjo Banda C et al. 2019 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Self-management is key to the control of glycaemia and prevention of complications in people living with diabetes. Many people living with diabetes in Malawi have poorly controlled glucose and they experience diabetes-related complications. This study aimed to assess diabetes self-management behaviours and to identify factors associated with it among people living with diabetes at Queen Elizabeth Central Hospital (QECH), Blantyre, Malawi.

Methods: This cross-sectional study recruited 510 adults attending a diabetes clinic at a teaching referral hospital in southern Malawi. The social cognitive theory was applied to identify factors associated with following all recommended self-management behaviours. Data on participants’ demographics, clinical history, diabetes knowledge, self-efficacy, outcome expectations, social support, environmental barriers and diabetes self-management were collected. Univariate and multivariate logistic regression analyses were conducted to identify factors associated with following all self-management behaviours.

Results: The mean age of participants was 53.6 (SD 13.3) years. Self-reported medication adherence within the last seven days was 88.6% (n=494); 77% reported being physically active for at least 30 minutes on more than three days in the previous seven days; 69% reported checking their feet every day and inspecting inside their shoes; 58% reported following a healthy diet regularly. Overall, only 33% reported following all the self-management behaviours regularly. Multiple logistic regression analysis showed that self-efficacy was the only social cognitive factor associated with following all the self-management practices (p < 0.001).

Conclusions: Participants in our study were not consistently achieving all self-management practices with dietary practices being the least adhered to behaviour by many. To improve self-management practices of people living with diabetes, current health education programs should not only aim at improving diabetes related knowledge but also self-efficacy. Adopting interventions that promote self-efficacy in diabetes patients such as exposure to role models, peer education, providing positive feedback, and counselling is recommended.

 diabetes education glycemic control physical activity healthy diet foot care social cognitive theory environmental factors social support New Partnership for Africa's Development Department for International Development, UK Government National Heart, Lung, and Blood Institute 5U24HL136791-01 Carnegie Corporation of New York B8606.R02 National Institutes of Health 5U24HL136791-01 African Academy of Sciences University of the Witwatersrand, Johannesburg B8606.R02 African Population and Health Research Center B8606.R02 Wellcome Trust 107768 Styrelsen för Internationellt Utvecklingssamarbete 54100029 This research was supported by the Consortium for Advanced Research Training in Africa (CARTA). CARTA is jointly led by the African Population and Health Research Center and the University of the Witwatersrand and funded by the Carnegie Corporation of New York (Grant No--B 8606.R02), Sida (Grant No:54100029), the DELTAS Africa Initiative (Grant No: 107768/Z/15/Z). The DELTAS Africa Initiative is an independent funding scheme of the African Academy of Sciences (AAS)’s Alliance for Accelerating Excellence in Science in Africa (AESA) and supported by the New Partnership for Africa’s Development Planning and Coordinating Agency (NEPAD Agency) with funding from the Wellcome Trust (UK) (Grant No: 107768/Z/15/Z) and the UK government, the statements made and views expressed are solely the responsibility of the fellow. MN and ASM are also supported by the NCD BRITE consortium. The Consortium is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under grant number 5U24HL136791-01. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

Diabetes mellitus significantly contributes to morbidity and mortality from non-communicable diseases in Malawi 1 . Diabetes is the ninth-leading cause for admissions in adult medical wards at Queen Elizabeth Central Hospital (QECH) in Blantyre, the largest public teaching hospital in Malawi 2 . The inpatient mortality for the people admitted due to diabetes at QECH is about 19% 2 . A previous survey at QECH by Cohen et al., conducted 10 years before the present study found that 74% of people living with diabetes had poorly controlled sugar levels and many suffered from diabetes related complications 3 . This previous survey also found that 45% of patients living with diabetes had poor dietary practices. Among the patients that were on insulin, about 22% had problems with proper injection technique 3 . However, Cohen et al. did not report on other self-management practices such as medication adherence, self-monitoring of blood glucose, foot care, smoking habits and exercises. Subsequent to the study by Cohen et al., clinical guidelines and protocols for the management of diabetes and nurse-led education classes for diabetes patients were introduced at QECH 3 . The nurse-led education classes offer lessons to people living with diabetes on lifestyle practices related to diet, exercises, medication adherence, smoking cessation, foot care, and management of symptoms generated by the disease to help keep diabetes under control and to prevent its complications.

Although the study by Cohen et al. had shown that people living with diabetes in Malawi had poorly controlled glucose 3 , little was known about their self-management behaviours especially regarding diet, exercising, self-monitoring of blood glucose, medication adherence, and foot care. Furthermore, since the introduction of the nurse-led diabetes education classes, no follow-up study was conducted to evaluate if there has been an impact on people’s diabetes self-management behaviours.

The conceptual framework guiding the study was adapted from social cognitive theory (SCT) by Albert Bandura 4 . Recent studies continue to show that the propositions made in the SCT on determinants of health behaviour remain valid to date, not only for diabetes but many other chronic conditions. A systematic review of randomized controlled trials of SCT-based interventions showed effectiveness in improving diet and exercise behaviour in cancer survivors 5 . The social cognitive theory outlines several factors that are key to the acquisition of knowledge and skills which influence health and wellbeing of individuals. Some key concepts of the social cognitive theory are self-efficacy, health knowledge, health goals, outcome expectations, and environmental impediments and facilitators 4, 6 . These key concepts are the factors that influence human action, motivation and wellbeing and hence are hypothesized to be associated with diabetes self-management for this present study 6 . The aim for this study was to assess the level of self-management and identify factors associated with practicing all self-management behaviours among adults living with diabetes at QECH. This is part of a larger study exploring self-management practices and experiences among people living with diabetes attending the QECH diabetes clinic.

 Methods

A cross-sectional design was used to collect demographics, clinical, diabetes knowledge, self-efficacy, outcome expectations, social support, environmental barriers and diabetes self-management data. Ethical approval for the study was granted by the College of Medicine Research and Ethics Committee (Ref: P.08/17/229). All participants provided written informed consent to participate in the study.

Inclusion and exclusion criteria

Recruitment of clients was done on diabetes clinic days at QECH. The diabetes clinic at QECH runs once a week. Clients were eligible to participate if they were aged 18 years and above, had clinically confirmed type 1 or type 2 diabetes mellitus for over six months and were available at the clinic between 9am and 1pm, the time when data were being collected. Clients were excluded if they had cognitive impairment or communication difficulties, had lived with diabetes for less than 6 months or were acutely ill. To avoid selection bias, systematic sampling was used and an invitation was made to every third person on the queue who met the recruitment criteria until the required sample size was met.

 Sample size considerations

It was assumed that the proportion of participants achieving good self-management could be 0.26 (26%), based on the previous study at QECH by Cohen et al. (2010) 3 . The proportion was estimated to be within 0.04 (4%), using the 95% confidence level a sample of 462 patients living with diabetes would be required. The final sample size of 510 was obtained after a 10% adjustment for refusals and to account for potential confounding factors (age, sex, type of diabetes and duration of diabetes since diagnosis).

 Data collection and instruments

Data were collected between November 2017 and May 2018 using an interviewer-administered questionnaire 7 . The questionnaire was administered by the first author and five other trained research assistant all of whom have a background in nursing. To mitigate social desirability bias whereby respondents tend to over-report healthy behaviours or under-report unhealthy behaviours, during recruitment the participants were informed of the anonymity of the data and how their participation or refusal to participate in the study would not affect their care at the diabetes clinic 8 . Furthermore, the researcher and research assistants did not wear nurses’ uniform during data collection to create a neutral environment. The questionnaire collected data on the participants’ demographics, clinical history, social cognitive theory constructs (diabetes knowledge, self-efficacy, outcome-expectations, environmental barriers to proper self-management and social support) and self-management and it is available as Extended data on Figshare 7 . Clinical data were extracted from participants’ health passport books and included weight, height, body mass index (BMI), blood pressure reading for that day, fasting blood glucose (FBG) reading for that day and for the client’s last two clinic visits, creatinine checked within the last 12 months, time since diabetes diagnosis, type of diabetes, type of treatment, diabetes complications, and if there were any comorbidities including HIV status.

To measure self-management, we adapted ten items from the Summary of Diabetes Self-Care Activities (SDSCA) measure and one item from the expanded version of the SDSCA developed by Toolbert et al. 9 . We used all four items of the SDSCA on diet. We dropped one of the two items on exercises as the participants who took part in the content validation of the questionnaire felt that the items were asking the same thing. The items on blood sugar testing were dropped as content experts who reviewed the tool felt that the questions were not applicable in the Malawian setting since many people may not have a personal glucometer. Instead we included a question that asked the participants if they have a glucometer at home. Reliability and validity of the SDSCA measure has been proved from previous studies with a high correlation with other scales of self-care 9 . The adapted SDSCA assessed level of self-care related to diet (four questions), exercise (one question), blood sugar testing (one question), foot care (two questions), smoking (two questions) and medication (one question). For each subscale, the respondent was asked to mention number of days they performed a particular activity in the past one week. Reverse scoring was done for the question on fat intake. For all the questions, a higher number of days indicated satisfactory self-management.

Diabetes knowledge was measured using items adapted from the Diabetes Knowledge Questionnaire (DKQ) 10 . We used all the 24 items of the DKQ to measure knowledge on causes, signs and symptoms, pathophysiology and treatment of diabetes. The tool had showed construct validity and reliability in a Mexican-American population with Cronbach’s coefficient α of 0.78 10 . For each item, the respondents answered either “yes”, “no” or “I don’t know”. The total number of correct answers was calculated at the end to obtain the knowledge score.

Self-efficacy is a person’s belief or judgement in their ability to accomplish specific acts 11 . The Self-efficacy for Diabetes Tool 12 was used to measure self-efficacy. All the eight items of the tool were used to asks of the respondents’ confidence to perform various diabetes self-management tasks on a scale of 1 to 10, where 1 was “not at all confident” and 10 was “totally confident” 13 . Scoring of the scale was based on the mean of at least six items with higher scores indicating higher self-efficacy. This tool had been used in previous studies with a Cronbach’s coefficient α of 0.85 and a test-retest validity of 0.8 13 .

Outcome expectations refers to a persons’ belief or anticipated result for doing a particular behaviour 6 . The Outcome expectations were assessed using items adapted from the multidimensional diabetes questionnaire 14 . We adapted all six items related to outcome expectations from the questionnaire to ask participants’ perceptions on the effects of performing particular self-care activities on their glucose control or prevention of diabetes related complications. The scores ranged from 1 (not at all important) to 10 (totally important). Scoring of the scale was based on the mean of the six items with higher scores showing more positive expectancies. A previous study assessing the validity of this test found a Cronbach’s coefficient α of 0.86 14 .

Environmental factors that were assessed were social support and barriers to self-management. Social support is a multidimensional construct that refers to “a network of family, friends, neighbours, and community members that is available in times of need to give psychological, physical, and financial help” 15 . Social support was assessed using a nine-item measure of social support with a five-point-Likert scale. The tool assessed availability of emotional, informational support, networking support and sources of social support 16 . This tool was adapted from the Medical Outcomes Study social support and the items showed reliability (Cronbach’s coefficient α ranging 0.74 to 0. 93) and construct validity 17 . Scoring was based on the frequencies of each item.

We used 24 items out of the 31 items of the environmental barriers to diabetes adherence tool 18 to assess environmental barriers of the participants. Six items on self-glucose monitoring were dropped during content validation of the tool to suit context as most people living with diabetes in Malawi do not have personal glucometers. One other item that stated “I feel sore and stiff” was dropped as participants in the content validation exercise of the Chichewa version felt that the item had the same meaning with the item that stated “I don’t feel well”.

 Data analysis

Data were entered into a Microsoft Access database then exported into Stata version 14.0 for cleaning and analysis. Descriptive statistics were used to show proportions and 95% confidence interval (CI) for categorical factors and mean and standard deviations (SD) for continuous factors that were normally distributed. Median and interquartile range (IQR) were calculated for factors that were not normally distributed. The outcome variable was self-management as measured by the SDSCA. Participants were categorized as having adequate self-management behaviour if they were adherent to all four self-management practices (diet, exercise, medication and foot care). There were only 14 people who were not on any diabetes medication, who were therefore excluded in the final analysis. Adherence to blood glucose self-monitoring was not assessed as only 12% of the participants had personal glucometers and all reported not to have measured themselves in the last seven days. Smoking status was not included as there were only three active smokers. Univariate logistic regression was done to investigate associations between demographics, clinical and social cognitive factors with the outcome variable. Chi-square (or Fisher’s exact) test and t-test (Wilcoxon rank sum test) were used for testing association between the binary outcome of self-management behaviour with categorical explanatory factors and continuous factors, respectively. Factors that showed association with adequate self-management at alpha 0.1 or less were included in the multivariate logistic regression model. Participants’ sex, diabetes type and duration of diabetes diagnosis were included in the multivariate logistic regression model because they were believed to be possible confounders. Associations were considered significant at alpha less than 0.05.

 Results Participant background

A total of 554 clients were selected and invited to participate in the study using systematic random sampling, of which 538 met the recruitment criteria and 28 refused to participate. In total, 510 consented to participate, representing a response rate of 95%. Overall, there were more females (82%). A total of 14 participants were excluded from the final analysis for having no data on medication adherence. Table 1 contains the demographic characteristics and in Table 2 are the clinical characteristics of the study participants.

Participants demographic characteristics at QECH, Malawi.
Demographic Characteristics Characteristic n * %
Sex Male 91 17.84
Female 419 82.16
Age, years Mean (SD) 53.65 (13.31)
Marital status Never married 14 2.75
Married 220 64.71
Divorced 39 7.65
Widowed 119 23.33
Separated 8 1.57
Education Never been to school 54 10.59
Primary school 249 48.82
Secondary school 180 35.29
Tertiary education 27 5.29
Occupation Employed 51 10
Farming 64 12.55
Small scale business 214 41.96
Unemployed 148 29
Retired 33 6.47

*Unless indicated. SD, standard deviation.

 Clinical characteristic of participants at QECH, Malawi.
Variable Characteristics Characteristic n * %
Duration 5 years or less 272 54.08
6 – 10 year 130 25.84
11 – 15 years 57 11.33
More than 15 years 44 8.75
Diabetes type Type 1 56 10.98
Type 2 443 86.86
Unknown 11 2.16
Treatment Insulin only 92 18.07
Oral agents 348 68.37
Insulin and oral agents 55 10.81
Diet and exercise only 14 2.75
Complications None 208 40.78
One 229 44.9
Two or more 73 14.31
HIV status Negative 406 79.61
Positive 76 14.90
Unknown 28 5.49
Comorbidities None 181 35.49
One 306 60
Two or more 23 4.51
BMI Underweight 26 5.27
Normal 152 30.83
Overweight 162 32.86
Obese 153 31.03
FBG Median (IQR) 171.37 (129.24–234.51)
Systolic BP Median (IQR) 131 (118–146)
Diastolic BP Median (IQR) 84 (75–91)

*Unless indicated. BMI, body mass index; BP: blood pressure; FBG: fasting blood glucose; IQR, interquartile range.

 Participant questionnaire responses

The median knowledge score on the diabetes knowledge questionnaire was 14 (IQR 12–16) with lowest knowledge scores being on causes of diabetes, importance of diet and exercising and recognition of hypoglycemia or hyperglycemia. The median self-efficacy score was 8.6 (IQR 7.5–9.5), and the participants had lower self-efficacy on eating evenly spaced meals regularly and exercising for at least 30 minutes three times a week. The median for outcome expectations score was 10 (IQR 10–10), suggesting that participants had positive expectations in following recommended self-management behaviours. The median social support score was 4.9 (IQR 2.9–5). The most commonly mentioned sources of social support were spouses and daughters. The median score for environmental barriers to self-care was 1.5 (IQR 1.3–1.8). Barriers to medication were infrequent with only 7% of the participants reporting encountering barriers at any point. Barriers to healthy diet and exercising were reported by 37% and 33% of the participants respectively.

Most participants reported taking their medication everyday as recommended (88.66%) and also reported being physically active for at least 30 minutes on three or more days per week (71.41%). Physical activities reported included walking or engaging in one’s daily duties. Daily foot care was reported by 69.23% of the participants. For the general diet, 57.28% of the participants reported a healthy diet 6–7 days per week. On the specific diet, none of the participants reported taking at least five portions of fruits and vegetables per day, while 48.53% reported not to have taken any high fat food on any day in the previous seven days. Participants were considered to have satisfactory self-management behaviour if they reported regular adherence to the general diet, exercising, foot care and medication intake. Only 33% of the participants were adherent to all the four self-management behaviours. Figure 1 shows the percentage of participants who reported adherence to specific self-management behaviours and all self-management behaviours. All responses are given as Underlying data 19 .

Percentage of participants reporting adherence to self-management behaviours. Univariate and multivariate analysis

To investigate the factors associated with adhering to all the self-management behaviours, univariate and multivariate regression analyses were done. The unadjusted logistic regression analyses showed that satisfactory self-management was associated with self-efficacy, social support, outcome expectations and diabetes barrier score (the results are shown in Table 3).

Associations between demographic, clinical and social cognitive theory constructs with satisfactory self-management.
Variable Characteristic Odds ratio 95% CI p-value
Sex Female 1.007 0.613–1.655 0.979
Age (in years) Every additional year 1.000 0.986–1.014
Marital status Never married Reference
Married 3.100 0.682–14.103 0.143
Divorced 3.36 0.656–17.212 0.146
Widowed 2.530 0.538–11.900 0.24
Separated 10 1.260–79.339 0.029
Education Never been to school Reference
Primary school 1.728 0.861–3.470 0.124
Secondary school 1.892 0.925–3.869 0.081
Tertiary education 0.877 0.270–2.848 0.827
Occupation Employed Reference
Farming 0.555 0.253–1.216 0.141
Small scale business 0.707 0.279–1.794 0.466
Unemployed 0.539 0.283–1.024 0.059
Retired 0.705 0.363–1.371 0.303
BMI Underweight 0.822 0.305– 2.210 0.697
Normal Reference
Overweight 1.367 0.844–2.213 0.204
Obese 1.283 0.786–2.094 0.319
Systolic BP every additional unit 1.002 0.994–1.011 0.556
Diastolic BP every additional unit 0.999 0.985– 1.014 0.932
FBG every additional unit 0.999 0.999– 1.001 0.271
Creatinine Every additional unit 1.362 0.895– 2.073 0.149
Duration 5 years or less Reference
6–10 year 1.034 0.657–1.627 0.886
11–15 years 1.387 0.764–2.515 0.282
More than 15 years 1.0344 0.520–2.059 0.923
Diabetes type Type 1 Reference
Type 2 0.796 0.447–1.418 0.439
Unknown 1.111 0.280– 4.399 0.881
Diabetes treatment Insulin only Reference
Oral agents 0.925 0.571–1.501 0.754
Insulin and oral agents 0.839 0.410– 1.715 0.63
Diet and exercise only omitted
Complications None Reference
One 0.968 0.648–1.447 0.875
Two or more 0.694 0.380–1.265 0.233
Comorbidities None Reference
One 0.969 0.654–1.437 0.876
Two or more 0.285 0.081–0.998 0.05
HIV status Negative Reference
Positive 1.303 0.783–2.167 0.309
Unknown 1 0.440–2.273 1
Knowledge Every additional score 0.960 0.894–1.030 0.416
Self-efficacy Every additional score 1.499 1.288–1.744 0.0001
Outcome expectations Every additional score 1.386 0.954– 2.013 0.087
Social support Every additional score 1.276 1.031– 1.579 0.025
Total barrier score Every additional score 0.536 0.321–0.893 0.017

CI, confidence interval; BMI, body mass index; BP, blood pressure; FBG, fasting blood glucose.

In the multivariate logistic regression model, we adjusted for age, sex, duration since diabetes diagnosis and type of diabetes. The results showed that self-efficacy was the only significant factor associated with satisfactory self-management (p < 0.001). For a one-unit increase in the self-efficacy score, the odds of having satisfactory self-management increase by 1.5 (CI 1.2 – 1.7).

 Discussion

This study applied the social cognitive theory to assess self-management behaviours and its associated factors among patients living with diabetes at an urban diabetes clinic in Blantyre, Malawi. Medication adherence was highest of all the self-management behaviours that were assessed. High rate of adherence to medication have also been reported in previous studies from the USA 9 , Ethiopia 20, 21 and rural Malawi 22 . Our results suggest that people living with diabetes attending the QECH diabetes clinic have fewer environmental barriers to medication adherence than to other self-care practices. The high levels of adherence to medication could also suggest that people living with diabetes prioritize medication intake over other self-management behaviours. Although medication adherence is associated with better glycemic control 22 , it should be accompanied with lifestyle modifications for better results 23, 24 .

A total of 71% of the participants also reported being physically active for at least 30 minutes three times a week as part of their daily work. The level of physical activity among the participants in our study was lower compared to findings from a population-based survey that was conducted in Malawi which reported that 90.5% adults were physically active 25 . Engaging in regular physical activity among people living with diabetes contributes to cardiorespiratory fitness, improved glycaemic control, decreased insulin resistance, improved blood lipid profile and improved blood pressure 23, 24, 26 . Our results suggested that the participants had low self-efficacy to exercise and frequently encountered barriers to exercising. This corresponds to findings from other studies from USA, Nigeria and Ethiopia, who also found lower rates of physical activity among people living with diabetes; this was attributed to low self-efficacy and high perceived barriers to physical activity 20, 21, 27, 28 .

Foot care was another self-management aspect practiced by most participants. Although foot care may not directly influence glycaemic control, it is an important self-management practice for the prevention of foot ulcers and leg amputations 24, 29 . People living with diabetes are prone to foot ulcers due to peripheral neuropathy which result from poor glycaemic control 23, 24 . In total, 69% of the participants in our study reported checking their feet daily and checking inside their shoes before wearing them every day. This contrasts with the findings of Assayed et al., at Mangochi District Hospital in Malawi, where only 17% of diabetic patients reported inspecting their feet regularly, and 15% did not wear shoes at all 22 . This observed difference between our study and that of Assayed et al. could be due to differences in settings and the quality of service delivery. Mangochi district is mostly rural and with a limited capacity of providing diabetes self-management education 22 . Although many patients reported daily foot care, it is however not adequate considering that most of them had peripheral neuropathy. Literature shows that QECH has a high number of people living with diabetes who present late with ulcers, which may result in limb amputations 30 .

Following a recommended healthy diet was the least regularly practiced self-management behaviour and corresponds with findings from a study that was conducted in the USA 28 . The recommended diet for people with diabetes mainly consists of foods that have low carbohydrate, low salt, whole grains, fruits and vegetables 23, 24, 31 . Additionally, a healthy diet restricts fats, sweetened foods or beverages, and recommends eating of small food potions spread out evenly throughout the day 23, 24, 31 . For their general diet, 57% of the participants reported following a healthy diet as recommended at least six days a week. The specific diet assessment showed that none of the participants were taking at least five portions of fruits and vegetables every day. This is similar to what was found in a population-based national survey conducted in Malawi, where fruit intake was on average two days per week 25 . Following a healthy diet plan can reduce glycated haemoglobin (HbA1C) levels by up to 2% and is protective from cardiovascular and non-cardiovascular disease mortality for people living with diabetes 23, 32 ; therefore, it should be encouraged.

Self-monitoring of blood glucose was not assessed as only 12% of the participants reported to have a glucometer at home. Nevertheless, lower rates of self-monitoring of blood glucose have been reported in previous studies conducted in sub-Saharan African countries like Tanzania 33 and Kenya 34 . Low rates of self-monitoring of blood glucose in people living with diabetes in Africa has been attributed to financial constraints 33, 35 . In contrast, studies conducted in high-income countries like France 36 , Sweden 33 and Italy 35 have reported regular self-monitoring of blood glucose among people living with diabetes. Regular monitoring of blood glucose is associated with good glycemic control 35 .

Overall, we found that only one-third (33%) of the participants were following all (diet, exercise, foot care and medication) the recommended self-management practices. Other studies have also found that most people living with diabetes do not follow all the recommended self-management practices. A study in Ethiopia, found that only 39% were following all the self-management practices 20 . A study in Mexico found that only 26% were following all the recommended self-care activities 37 . In another study by Zulman et al. in USA, only 26% reported performing four or five of the five self-management behaviours which they assessed 28 . Failure to follow all recommended self-management behaviours may be due to the fact that each self-management behaviour has different barriers and requires different knowledge, skills and motivation 38 .

We found that self-efficacy was the only significant (p < 0.001) social cognitive theory factor associated with following all self-management behaviours. Many studies have also found self-efficacy as a predictor to all self-management behaviours independently or collectively 27, 39– 43 . The social cognitive theory suggests that people with high self-efficacy set high goals for themselves, are more positive minded and have better analytical skills 6 . Additionally, studies have shown that diabetes self-efficacy is also associated with other predictors of diabetes self-management such as health literacy, health related quality of life and social support 43– 45 .

The other social cognitive theory constructs (outcome expectations, social support, environmental barriers and knowledge) showed no statistically significant association with satisfactory self-management. Social support and environmental barriers to self-care scores were however associated with satisfactory self-management in the univariate analysis (p < 0.05) but lost their significance in the multivariate logistic regression model. There are mixed findings on the association between outcome expectations, social support, knowledge and environmental barriers as predictors of one or more self-management behaviours. Some studies have reported an association of any of these with self-management 27, 46– 48 while others reported no associations 49– 51 . Self-efficacy is, however, the main factor that regulates all the other constructs of the social cognitive theory as it influences feelings, motivation, thoughts, expectations and goals 52 . Self-efficacy is also associated with other social cognitive constructs such as social support 53 ; therefore, more studies are required to explore further the relationship of the other social cognitive theory constructs with each other and diabetes self-management.

This study had several limitations. One of the limitations was that the participants were predominantly female. However, there were no statistically significant differences in following of self-management behaviours between males and females. Another limitation was that the study was hospital-based and recruited participants from one health facility only. Participants attending the clinic may be more compliant to self-management behaviours than those who do not come to the clinic. Additionally, generalizability of the findings is limited to central hospitals or health facilities of similar nature. Although the study identified factors associated with poor dietary behaviours, causality cannot be ascertained. Experimental studies are needed to identify locally appropriate and acceptable interventions that can improve self-efficacy in diet and all other self-management behaviours.

 Conclusion

The findings of this study show that people living with diabetes attending QECH diabetes clinic were not consistently following all the recommended self-management practices. Dietary practices were the least adhered to self-management behaviour compared to medication, foot care and exercising. Management protocols and guidelines for people living with diabetes at QECH should therefore include interventions aimed at improving self-efficacy such as exposure to role models, peer education, providing positive feedback, and counselling. We also noted that most of the people living with diabetes lacked access to resources that enabled them to perform self-monitoring of blood glucose. We recommend availability of blood glucose monitoring devices at primary health care level and even to all individuals living with diabetes to allow regular monitoring of blood glucose, which is necessary for the adjustment of medication, diet and exercise intensity.

 Data availability Underlying data

Figshare: Diabetes self-management and social cognitive factors. https://doi.org/10.6084/m9.figshare.9757076.v1 19 .

This project contains answers to each question from each respondent. The first row contains the question number from the questionnaire (see Extended data) to which the answer pertains.

 Extended data

Figshare: Factors associated with diabetes self-management questionnaire. https://doi.org/10.6084/m9.figshare.9757115.v1 7 .

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

 Acknowledgements

We wish to acknowledge the study participants, research assistants and all staff working at QECH diabetes clinic for their support during the study.

 Malawi Government: National Action Plan for Prevention and Management of Non-Communicable Diseases in Malawi (2012-2016). Ministry of Health;2013. Allain TJ Aston S Mapurisa G : 127 A Descriptive Study of Patterns of disease and Clinical Outcomes in Older Adults Admitted to Medical Wards in a Central Hospital in Malawi. Age Ageing. 2014;43(suppl_1):i35. 10.1093/ageing/afu046.10 Cohen DB Allain TJ Glover S : A survey of the management, control, and complications of diabetes mellitus in patients attending a diabetes clinic in Blantyre, Malawi, an area of high HIV prevalence. Am J Trop Med Hyg. 2010;83(3):575–81. 20810823 10.4269/ajtmh.2010.10-0104 2929054 Bandura A : Health promotion from the perspective of social cognitive theory. Psychol Health. 1998;13(4):623–49. 10.1080/08870449808407422 Stacey FG James EL Chapman K : A systematic review and meta-analysis of social cognitive theory-based physical activity and/or nutrition behavior change interventions for cancer survivors. J Cancer Surviv. 2015;9(2):305–38. 25432633 10.1007/s11764-014-0413-z 4441740 Bandura A : Health promotion by social cognitive means. Heal Educ Behav. 2004;31(2):143–64. 15090118 10.1177/1090198104263660 Kwanjo Banda C : Factors associated with diabetes self-management questionnaire.2019. http://www.doi.org/10.6084/m9.figshare.9757115.v1 King MF Bruner GC : Social Desirability Bias: A Neglected Aspect of Validity Testing. Psychol Mark. 2000;17(2):79–103. 10.1002/(SICI)1520-6793(200002)17:2<79::AID-MAR2>3.0.CO;2-0 Toobert DJ Hampson SE Glasgow RE : The summary of diabetes self-care activities measure: results from 7 studies and a revised scale. Diabetes Care. 2000;23(7):943–50. 10895844 10.2337/diacare.23.7.943 Garcia AA Villagomez ET Brown SA : The Starr County Diabetes Education Study: development of the Spanish-language diabetes knowledge questionnaire. Diabetes Care. 2001;24(1):16–21. 11194219 10.2337/diacare.24.1.16 Bandura A : Self-Efficacy. Encycl Hum Behav. 1994;4:71–81. 10.1002/9780470479216.corpsy0836 Stanford Patient Education Research Center. Self-Efficacy for Diabetes. 2009;1–2. Lorig K Ritter PL Villa FJ : Community-based peer-led diabetes self-management: A randomized trial. Diabetes Educ. 2009;35(4):641–51. 19407333 10.1177/0145721709335006 Talbot F Nouwen A Gingras J : The assessment of diabetes-related cognitive and social factors: the Multidimensional Diabetes Questionnaire. J Behav Med. 1997;20(3):291–312. 9212382 10.1023/A:1025508928696 National Cancer Institute: NCI Definitions of Cancer terms.n.d. (accessed April 4, 2019). Reference Source Diabetes Initiative: Advancing Diabetes Self Management Okland 2005 n.d. (accessed April 4, 2019). Reference Source Sherbourne CD Stewart AL : The MOS social support survey. Soc Sci Med. 1991;32(6):705–14. 2035047 10.1016/0277-9536(91)90150-b Glasgow RE : Social-environmental factors in diabetes: barriers to diabetes self-care. In: Bradley C editor. Handb Psychol diabetes a Guid to Psychol Meas diabetes Res Pract, Routledge;2013. Reference Source Kwanjo Banda C : Diabetes self-management and social cognitive factors.csv.2019. http://www.doi.org/10.6084/m9.figshare.9757076.v1 Ayele K Tesfa B Abebe L : Self care behavior among patients with diabetes in harari, eastern ethiopia: the health belief model perspective. PLoS One. 2012;7(4):e35515. 22530039 10.1371/journal.pone.0035515 3328462 Bonger Z Shiferaw S Tariku EZ : Adherence to diabetic self-care practices and its associated factors among patients with type 2 diabetes in Addis Ababa, Ethiopia. Patient Prefer Adherence. 2018;12:963–970. 29922042 10.2147/PPA.S156043 5995298 Assayed AA Muula AS Nyirenda MJ : The quality of care of diabetic patients in rural Malawi: A case of Mangochi district. Malawi Med J. 2014;26(4):109–14. 26167259 4325344 Amod A Ascott-Evans BH Berg GI : The 2012 SEMDSA Guideline for the Management of Type 2 Diabetes (Revised). J Endocrinol Metab Diabetes South Africa. 2012;17(2):S1–95. Reference Source American Diabetes Association: Standards of Medical Care in Diabetes-2019 Abridged for Primary Care Providers. Clin Diabetes. 2019;37(1):11–34. 30705493 10.2337/cd18-0105 6336119 Msyamboza KP Ngwira B Dzowela T : The burden of selected chronic non-communicable diseases and their risk factors in Malawi: nationwide STEPS survey. PLoS One. 2011;6(5):e20316. 21629735 10.1371/journal.pone.0020316 3100352 Peirce NS : Diabetes and exercise. Br J Sports Med. 1999;33(3):161–72; quiz 172–3, 222. 10378067 10.1136/bjsm.33.3.161 1756173 Adeniyi A Idowu O Ogwumike O : Comparative influence of self-efficacy, social support and perceived barriers on low physical activity development in patients with type 2 diabetes, hypertension or stroke. Ethiop J Health Sci. 2012;22(2):113–9. 22876075 3407834 Zulman DM Rosland AM Choi H : The influence of diabetes psychosocial attributes and self-management practices on change in diabetes status. Patient Educ Couns. 2012;87(1):74–80. 21840149 10.1016/j.pec.2011.07.013 3229832 Abbott CA Carrington AL Ashe H : The North-West Diabetes Foot Care Study: incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort. Diabet Med. 2002;19(5):377–84. 12027925 10.1046/j.1464-5491.2002.00698.x Kasiya MM Mang'anda GD Heyes S : The challenge of diabetic foot care: Review of the literature and experience at Queen Elizabeth Central Hospital in Blantyre, Malawi. Malawi Med J. 2017;29(2):218–23. 28955436 10.4314/mmj.v29i2.26 5610299 World Health Organization: Global Report on Diabetes.Isbn.2016;978:88. Reference Source Nöthlings U Schulze MB Weikert C : Intake of vegetables, legumes, and fruit, and risk for all-cause, cardiovascular, and cancer mortality in a European diabetic population. J Nutr. 2008;138(4):775–81. 18356334 10.1093/jn/138.4.775 Smide B Ekman L Wikblad K : Diabetes self-care and educational needs in Tanzanian and Swedish diabetic patients: a cross-cultural study. Trop Doct. 2002;32(4):212–6. 12405300 10.1177/004947550203200410 Wambui Charity K Kumar AMV Hinderaker SG : Do diabetes mellitus patients adhere to self-monitoring of blood glucose (SMBG) and is this associated with glycemic control? Experiences from a SMBG program in western Kenya. Diabetes Res Clin Pract. 2016;112:37–43. 26655019 10.1016/j.diabres.2015.11.006 Mannucci E Antenore A Giorgino F : Effects of Structured Versus Unstructured Self-Monitoring of Blood Glucose on Glucose Control in Patients With Non-insulin-treated Type 2 Diabetes: A Meta-Analysis of Randomized Controlled Trials. J Diabetes Sci Technol. 2018;12(1):183–189. 28697625 10.1177/1932296817719290 5761981 Guerci B Benhamou PY Durain D : [Self-monitoring of blood glucose in France: data from a national survey]. Sante Publique. 2017;29(2):229–40. 28737342 Lerman I Lozano L Villa AR : Psychosocial factors associated with poor diabetes self-care management in a specialized center in Mexico City. Biomed Pharmacother. 2004;58(10):566–70. 15589064 10.1016/j.biopha.2004.09.003 Al-Khawaldeh OA Al-Hassan MA Froelicher ES : Self-efficacy, self-management, and glycemic control in adults with type 2 diabetes mellitus. J Diabetes Complications. 2012;26(1):10–6. 22226484 10.1016/j.jdiacomp.2011.11.002 Kadirvelu A Sadasivan S Ng SH : Social support in type II diabetes care: a case of too little, too late. Diabetes Metab Syndr Obes. 2012;5:407–17. 23226028 10.2147/DMSO.S37183 3514066 King DK Glasgow RE Toobert DJ : Self-efficacy, problem solving, and social-environmental support are associated with diabetes self-management behaviors. Diabetes Care. 2010;33(4):751–3. 20150299 10.2337/dc09-1746 2845021 Reisi M Mostafavi F Javadzade H : Impact of Health Literacy, Self-efficacy, and Outcome Expectations on Adherence to Self-care Behaviors in Iranians with Type 2 Diabetes. Oman Med J. 2016;31(1):52–9. 26813680 10.5001/omj.2016.10 4720937 Amer FA Mohamed MS Elbur AI : Influence of self-efficacy management on adherence to self-care activities and treatment outcome among diabetes mellitus type 2. Pharm Pract (Granada). 2018;16(4):1274. 30637026 10.18549/PharmPract.2018.04.1274 6322981 Osborn CY Cavanaugh K Wallston KA : Self-efficacy links health literacy and numeracy to glycemic control. J Health Commun. 2010;15 Suppl 2:146–58. 20845200 10.1080/10810730.2010.499980 3082364 Mayberry LS Rothman RL Osborn CY : Family members’ obstructive behaviors appear to be more harmful among adults with type 2 diabetes and limited health literacy. J Health Commun. 2014;19 Suppl 2:132–43. 25315589 10.1080/10810730.2014.938840 4199389 Jahanlou AS Alishan Karami N : The effect of literacy level on health related-quality of life, self-efficacy and self-management behaviors in diabetic patients. Acta Med Iran. 2011;49(3):153–8. 21681702 Wang CM Inouye J Davis J : Diabetes knowledge and self-management effects on physiological outcomes in type 2 diabetes. Nurs Forum. 2013;48(4):240–7. 24188435 10.1111/nuf.12037 Huang J Liu Y Zhang Y : [Correlation between self-management and knowledge of and attitude to diabetes in type 2 diabetic patients in Changsha]. Zhong Nan Da Xue Xue Bao Yi Xue Ban. 2013;38(2):176–81. 23456075 10.3969/j.issn.1672-7347.2013.02.011 Glasgow RE Toobert DJ Gillette CD : Psychosocial Barriers to Diabetes Self-Management and Quality of Life. Diabetes Spectr. 2001;14(1):33–41. 10.2337/diaspect.14.1.33 Chlebowy DO Garvin BJ : Social support, self-efficacy, and outcome expectations: impact on self-care behaviors and glycemic control in caucasian and African American adults with type 2 diabetes. Diabetes Educ. 2006;32(5):777–86. 16971711 10.1177/0145721706291760 Glasgow RE Edwards LL Whitesides H : Reach and effectiveness of DVD and in-person diabetes self-management education. Chronic Illn. 2009;5(4):243–9. 19933245 10.1177/1742395309343978 Stark Casagrande S Ríos Burrows N Geiss LS : Diabetes knowledge and its relationship with achieving treatment recommendations in a national sample of people with type 2 diabetes. Diabetes Care. 2012;35(7):1556–65. 22498806 10.2337/dc11-1943 3379593 Bandura A : Human Agency in Social Cognitive Theory. Am Psychol. 1989;44(9):1175–84. 2782727 10.1037/0003-066x.44.9.1175 Mansyur CL Rustveld LO Nash SG : Hispanic Acculturation and Gender Differences in Support and Self-Efficacy for Managing Diabetes. Diabetes Educ. 2016;42(3):315–24. 27033722 10.1177/0145721716640905 10.21956/aasopenres.14076.r28577 Reviewer response for version 1 Brew-Sam Nicola 1 Referee https://orcid.org/0000-0001-7544-4499 Department of Health Services Research and Policy, Research School of Population Health, Australian National University, Canberra, Australia

Competing interests: No competing interests were disclosed.

 24 5 2021 Copyright: © 2021 Brew-Sam N 2021 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Thank you very much for the interesting study, I think it is well conducted with some minor weaknesses in the manuscript. Please find my suggestions below:

 

The gender distribution could be added to the abstract.

n=494 in the abstract's result section only applies for one result, and the other results refer to N=510? This is a bit unclear. I would suggest removing n=494 from the abstract.

Percentages should be reported consistently without decimals 88.6% = 89%

"Although the study by Cohen et al. had shown that people living with diabetes in Malawi had poorly controlled glucose, little was known about their self-management behaviours especially regarding diet, exercising, self-monitoring of blood glucose, medication adherence, and foot care" - this is a repetition of the first paragraph.

"A systematic review of randomized controlled trials of SCT-based inter- ventions showed effectiveness in improving diet and exercise behaviour in cancer survivors" - I would suggest to report another example on diabetes here, the cancer example doesn't fully fit. There is plenty of diabetes research focusing on SCT.

"A cross-sectional design" - please specify which data collection methods were used (standardised face-to-face survey?).

"95% confidence level" - better: confidence interval.

"The proportion was estimated to be within 0.04 (4%), using the 95% confidence level a sample of 462 patients living with diabetes would be required" - I don't understand this, please explain.

"Interviewer-administered questionnaire" - is that a standardized face-to-face survey? Or semi-structured interviews?

Page 4 Toobert, not "Toolbert".

"Measure has been proved from previous studies" - proven.

"For all the questions, a higher number of days indicated satisfactory self-management" - what is a higher number? Which scale values were rated as satisfactory? Please explain this, as it is relevant for the results.

"The tool had showed construct validity" - had shown.

"Cronbach’s coefficient alpha" - better: Cronbach’s alpha coefficient.

"For doing a particular behaviour" - better: executing.

"Univariate logistic regression was done to investigate" - better: used.

"The unadjusted logistic regression analyses showed that satisfactory self-management was associ- ated with self-efficacy, social support, outcome expectations and diabetes barrier score (the results are shown in Table 3)." - outcome expectations is not significant in Table 3, why is it still included in these results? Please explain.

Include a sub-heading for "study limitations".

Check paper again for spelling and grammar, there are some mistakes as outlined above.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

Health service research, diabetes research, mHealth/eHealth

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Kwanjo Banda Chimwemwe University of Malawi College of Medicine, Malawi

Competing interests: No competing interests were disclosed.

 10 12 2021

Approved With Reservations

Thank you very much for the interesting study, I think it is well conducted with some minor weaknesses in the manuscript. Please find my suggestions below:

The gender distribution could be added to the abstract.

The gender distribution has been added in the abstract

n=494 in the abstract's result section only applies for one result, and the other results refer to N=510? This is a bit unclear. I would suggest removing n=494 from the abstract.

N = 494 has been removed. This number was used because 16 people with diabetes who were recruited in the study were not on diabetes medication.

Percentages should be reported consistently without decimals 88.6% = 89%

Thank you. All percentages have been rounded up to whole numbers and decimals removed

"Although the study by Cohen et al. had shown that people living with diabetes in Malawi had poorly controlled glucose, little was known about their self-management behaviours especially regarding diet, exercising, self-monitoring of blood glucose, medication adherence, and foot care" - this is a repetition of the first paragraph.

The sentence in the previous paragraph has been deleted.

"A systematic review of randomized controlled trials of SCT-based inter- ventions showed effectiveness in improving diet and exercise behaviour in cancer survivors" - I would suggest to report another example on diabetes here, the cancer example doesn't fully fit. There is plenty of diabetes research focusing on SCT.

The cancer article has been removed and replaced with one reporting SCT in diabetes.

"A cross-sectional design" - please specify which data collection methods were used (standardised face-to-face survey?).

The data collection methods have been specified as follows: “A cross-sectional design using standardised face-to-face survey was conducted” 

"95% confidence level" - better: confidence interval.

Thank you. Confidence level has been changed to confidence interval

"The proportion was estimated to be within 0.04 (4%), using the 95% confidence level a sample of 462 patients living with diabetes would be required" - I don't understand this, please explain.

This has been rephrased to add clarity:

To determine the true proportion with satisfactory self-management at the 95% confidence level and at 4% level of precision, a minimum sample size of 462 people living with diabetes was required .

The formula used to calculate the sample size has been given.

"Interviewer-administered questionnaire" - is that a standardized face-to-face survey? Or semi-structured interviews?

Yes- it is standardized face-to-face survey. “Interviewer administered questionnaire” has been replaced with “standardized face-to-face survey for clarity.

Page 4 Toobert, not "Toolbert".

The name has been corrected from Toolbert to Toobert.

"Measure has been proved from previous studies" - proven.

This has been corrected

"For all the questions, a higher number of days indicated satisfactory self-management" - what is a higher number? Which scale values were rated as satisfactory? Please explain this, as it is relevant for the results

Thank you. This has been explained as follows: “Self-management was considered satisfactory if a person reported following the recommended practices related to diet, medication, and foot care on all days in the past seven days, and being active for at least 30 minutes on three days in the past seven day.

"The tool had showed construct validity" - had shown.

This has been corrected

"Cronbach’s coefficient alpha" - better: Cronbach’s alpha coefficient.

This has been corrected

"For doing a particular behaviour" - better: executing.

“Executed” used instead of “doing”

"Univariate logistic regression was done to investigate" - better: used.

“done” replaced with “used”

"The unadjusted logistic regression analyses showed that satisfactory self-management was associ- ated with self-efficacy, social support, outcome expectations and diabetes barrier score (the results are shown in Table 3)." - outcome expectations is not significant in Table 3, why is it still included in these results? Please explain.

Thank you for noting this. Outcome expectations were not associated with self-management as shown in Table 3. This has been corrected in the text

Include a sub-heading for "study limitations".

This has been included

Check paper again for spelling and grammar, there are some mistakes as outlined above.

Thank you. This has been done.

 

Is the work clearly and accurately presented and does it cite the current literature?

Yes

Is the study design appropriate and is the work technically sound?

 Yes

Are sufficient details of methods and analysis provided to allow replication by others?

 Partly

If applicable, is the statistical analysis and its interpretation appropriate?

 Partly

Are all the source data underlying the results available to ensure full reproducibility?

 Yes

Are the conclusions drawn adequately supported by the results?

 Yes

Thank you.

More details have been added as described above.

This has been improved as suggested above.

Thank you.

 10.21956/aasopenres.14076.r27803 Reviewer response for version 1 Petersen Fazlyn 1 Referee https://orcid.org/0000-0003-4592-2861 Department of Information Systems, University of the Western Cape, Cape Town, South Africa

Competing interests: No competing interests were disclosed.

 26 8 2020 Copyright: © 2020 Petersen F 2020 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Is the work clearly and accurately presented and does it cite the current literature? Partly:

“The inpatient mortality for the people admitted due to diabetes at QECH is about 19%” - can the actual value not be presented instead?

“A previous survey at QECH by Cohen  et al., conducted 10 years before the present study” - more recent studies should be used. 

Why is SCT for cancer being used when there is available literature for SCT and diabetes?

There is more recent literature available on diabetes self-management and SCT such as Ghoreishi et al. (2019 1 ), Bourne et al. (2020 2 ), Thojampa and Sarnkhaowkhom (2019 3 ) and Mirzaei-Alavijeh and Jalilian (2019 4 ).

The link between SCT and diabetes self-management could be better explained by using the references above, among others. 

 Is the study design appropriate and does the work have academic merit?

The work does have merit but I feel that SDSCA should be introduced in the last paragraph of the introduction. "Some key concepts of the social cognitive theory are self-efficacy, health knowledge, health goals, outcome expectations, and environmental impediments and facilitators". The use of SDSCA is not explicit in this sentence. 

 Are sufficient details of methods and analysis provided to allow replication by others?

Inclusion and exclusion criteria are well defined.

Sample size considerations - why is a 10-year-old study used as a basis instead of a sample size calculator like Raosoft?  http://www.raosoft.com/samplesize.html

It would be clearer to see a research model diagrammatically presented.

 If applicable, is the statistical analysis and its interpretation appropriate?

This tool had been used in previous studies with a Cronbach’s coefficient α of 0.85 and a test-retest validity of 0.8". As this survey is a combination of other surveys with adaptations made, I would recommend that Construct Reliability and Validity and Discriminant Validity be tested for the final survey used. At least Cronbach alpha should have been tested in this study too.

I find Table 3 rather difficult to understand. Why are there p-values shown for demographic variables? I assume that these are for testing the associations. 

I use structured equation modelling to test research models and hypotheses so this method is not very clear to me. "…social cognitive theory constructs (outcome expectations, social support, environmental barriers and knowledge) showed no statistically significant association with satisfactory self-management." - where are these hypotheses represented? I was expecting to see an independent and dependent variable in Table 3 to better understand the p-values.

It mentions "causality cannot be ascertained" but this needs to be clarified.

Limitations - "However, there were no statistically significant differences in following of self-management behaviours between males and females" - this statement is missing a reference.

 Are all the source data underlying the results available to ensure full reproducibility? Yes, the survey and results are available.

Are the conclusions drawn adequately supported by the results? Yes.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

The use of technology to improve diabetes self-management.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References : Self-care behaviors in patients with type 2 diabetes: Education intervention base on social cognitive theory. Diabetes Metab Syndr .13(3) : 10.1016/j.dsx.2019.04.045 2049-2056 31235135 10.1016/j.dsx.2019.04.045 : Mapping behavior change techniques to characterize a social cognitive theory informed physical activity intervention for adults at risk of type 2 diabetes mellitus. Transl Behav Med .2020;10(3) : 10.1093/tbm/ibz008 705-715 30715534 10.1093/tbm/ibz008 : The Social Cognitive Theory with Diabetes: Discussion. International Journal of Caring Sciences .2019;12(2) :1251 Reference source : A psychometric analysis of the socio-cognitive determinants questionnaire of type 2 diabetes prevention among a group at risk. Iranian Journal of Endocrinology and Metabolism .2019;20(5) :224-234 Reference source Kwanjo Banda Chimwemwe University of Malawi College of Medicine, Malawi

Competing interests: No competing interests were disclosed.

 10 12 2021

Is the work clearly and accurately presented and does it cite the current literature? Partly:

“The inpatient mortality for the people admitted due to diabetes at QECH is about 19%” - can the actual value not be presented instead?

The sentence has been rephrased and reads: The inpatient mortality for the people admitted due to diabetes at QECH is 19%.

 

“A previous survey at QECH by Cohen  et al., conducted 10 years before the present study” - more recent studies should be used. 

Noted. However there was no available literature on recent studies looking at self-management or glycemic control at QECH other than the Cohen et al. study.  This has been highlighted in the paper

 

Why is SCT for cancer being used when there is available literature for SCT and diabetes?

The reference has been replaced with one focusing on diabetes mellitus and STC.

 

There is more recent literature available on diabetes self-management and SCT such as Ghoreishi et al. (2019 1 ), Bourne et al. (2020 2 ), Thojampa and Sarnkhaowkhom (2019 3 ) and Mirzaei-Alavijeh and Jalilian (2019 4

Thank for the suggestions. These references have been included.

 

The link between SCT and diabetes self-management could be better explained by using the references above, among others. 

Noted. This has been done

Is the study design appropriate and does the work have academic merit?

 

The work does have merit but I feel that SDSCA should be introduced in the last paragraph of the introduction. "Some key concepts of the social cognitive theory are self-efficacy, health knowledge, health goals, outcome expectations, and environmental impediments and facilitators". The use of SDSCA is not explicit in this sentence. 

Thank you. A sentence has been added in the introduction paragraph to introduce the SDSCA.

Are sufficient details of methods and analysis provided to allow replication by others?

 

Inclusion and exclusion criteria are well defined.

Thank you.

 

Sample size considerations - why is a 10-year-old study used as a basis instead of a sample size calculator like Raosoft?  http://www.raosoft.com/samplesize.html

Noted. We will consider using the given site in future. We used the old study because that was the only available local study to help us determine the proportion estimated to have good self-management at the time of the study.

 

It would be clearer to see a research model diagrammatically presented.

A research model has been included.

If applicable, is the statistical analysis and its interpretation appropriate?

 

This tool had been used in previous studies with a Cronbach’s coefficient α of 0.85 and a test-retest validity of 0.8". As this survey is a combination of other surveys with adaptations made, I would recommend that Construct Reliability and Validity and Discriminant Validity be tested for the final survey used. At least Cronbach alpha should have been tested in this study too.

Thank you for that recommendation. Cronbach alpha for this study has been calculated and included in the paper.

 

I find Table 3 rather difficult to understand. Why are there p-values shown for demographic variables? I assume that these are for testing the associations. 

Table 3 has been simplified to include the association of self-management and social cognitive constructs (knowledge, self-efficacy, outcome expectations, social support and barriers) only . P values are used because the table shows results of univariate logistic regression analysis that tested independent associations between predictor variables (demographic characteristics, clinical factors and social cognitive constructs).

 

I use structured equation modelling to test research models and hypotheses so this method is not very clear to me. "…social cognitive theory constructs (outcome expectations, social support, environmental barriers and knowledge) showed no statistically significant association with satisfactory self-management." - where are these hypotheses represented? I was expecting to see an independent and dependent variable in Table 3 to better understand the p-values.

We used logistic regression analysis. The dependent variable was self-management dichotomized as satisfactory or unsatisfactory. Based on the conceptual framework guiding the study, our hypothesis was that the social cognitive theory constructs are associated with self-management. We had three categories of independent variables (1-demographic characteristics, 2- clinical characteristics and 3- social cognitive theory variables). The first step of the analysis was to test the independent associations between each of the independent variables and the dependent variable. Table 3 shows the results of the univariate logistic regression analysis, which shows the association between each of the dependent variable with self-management.

The second step was the multivariate regression analysis where independent variables that showed association with the dependent variable (self-management) were included.

 

It mentions "causality cannot be ascertained" but this needs to be clarified.

The sentence has been rephrased as follows: “Since this was a cross-sectional study, we were only able to identify factors associated with diabetes self-management and not the causes”

 

Limitations - "However, there were no statistically significant differences in following of self-management behaviours between males and females" - this statement is missing a reference.

This sentence has been rephrased as follows: However, at univariate analysis, we found no statistically significant differences in following of self-management behaviours between males and females.

Are all the source data underlying the results available to ensure full reproducibility? Yes, the survey and results are available.

Thank you

Are the conclusions drawn adequately supported by the results? Yes.

Thank you

 

Is the work clearly and accurately presented and does it cite the current literature?

Partly

Thank you. This has been improved and literature updated.

Is the study design appropriate and is the work technically sound?

Partly

This has been improved

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

This has been improved.

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

This has been improved and more details of the analysis are given.

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Are the conclusions drawn adequately supported by the results?

 Yes

Thank you.