Diabetes mellitus (DM) is a syndrome characterized by chronic hyperglycemia due to deficient production of insulin or ineffectiveness in its utilization by the cells, leading to abnormally high levels of glucose in circulation (International Diabetes Federation, 2019). It is the most common metabolic disease and has been recognized as a clinical syndrome since ancient times (Marianna et al., 2016). The development of the disorder is linked to the interaction between inherited and acquired factors notably the frequent consumption of easily accessible highly processed energy-dense fast food and persistent physical inactivity (Adeghate et al., 2006).

Currently, diabetes is identified as ubiquitous and crippling (Kumar et al., 2013) and the number of individuals with the disorder is increasing at an alarming rate. In Africa, the number is expected to increase from 19 million cases in 2020 to 47 million by 2045 with Nigeria, the most populous country in the continent, accounting for almost three million adult cases (International Diabetes Federation, 2021). As is the case with other parts of the world, Nigeria is likely to continue to experience an increase in the number of individuals with the disease, which will escalate the already existing burden.

In addition to the associated symptoms of the disease (frequent urination, increased thirst and increased hunger) (World Health Organization, 2013) which can overwhelm patients, serious long-term complications comprising stroke, coronary heart disease, chronic kidney disease, damage to the eyes and foot ulceration causing further suffering (World Health Organization, 2013). Diabetes can also negatively impact a wide range of key human physiological processes including sleep, energy conservation and immune function (Beccuti and Pannain, 2011; Grandner et al., 2011; Zizi et al., 2010). These important physiological processes, in particular sleep, appear to have a bi-directional relationship with the occurrence of diabetes. As an essential biological function, sleep is a critical determinant of overall health and survival (Rechtschaffen et al., 1989). Optimal levels of sleep have been shown to positively impact memory, learning, emotion, removal of toxin at cellular level and cardiometabolic function (Cincin et al., 2015; Davies et al., 2014; Frank et al., 2013; Xie et al., 2013). Concerns are mounting with respect to inadequate sleep (a significant part of modern lifestyle) contributing to the development of diabetes (Cappuccio et al., 2010; Kawakami et al., 2004; Öztürk et al., 2015). On the other hand, individuals with diabetes have reported a reduction in sleep quality, increased frequency of insomnia, excessive daytime sleepiness and higher sleep medication use (Skomro et al., 2001). Frequently, the precursors of sleep disturbances in diabetes are complications such as peripheral neuropathy, pain and polyuria, obstructive sleep apnea, restless leg syndrome and nocturnal hypoglycaemia (Surani, 2015), and these have been related to a heightened risk of cardiovascular complications with increased risk of death (Ayas et al., 2003). Thus, chronic sleep of short duration or low quality, known as poor sleep, is associated with compromise in both physical and mental health and can impact an individual’s ability to perform physical activity (PA) which is a significant adjunct in diabetes management.

Physical activity is defined as any movement (such as walking, running, dancing, biking, performing household chores, swimming, exercising, and engaging in sports activities) that uses skeletal muscles and requires more energy than resting (Caspersen et al., 1985). PA can improve sleep quality (Dolezal et al., 2017; Wang and Boros, 2019), while sleep restriction may negatively impact physical activity (Bromley et al., 2012) both of which may adversely affect quality of life.

Quality of life (QoL) is a significant multi-component concept that includes physical and socio-emotional aspects of an individual’s life. It is routinely used in research as a primary endpoint to indicate health status and monitor the impact of interventions in various chronic diseases including diabetes (Maharaj and Nuhu, 2015). Decreased quality of life adversely impacts not only the happiness and satisfaction of persons with DM, but also their engagement in socio-economic activities and compliance to treatment (Snel et al., 2012). Research evidence has proven that QoL for patients with DM is lower than that of their non-diabetic counterparts and that increases in levels of PA can help to offset this (Çolak et al., 2015; Maharaj and Nuhu, 2015). Factors such as advanced age, use of insulin, diabetes complications, inadequate sleep, psychosocial factors and so forth may interfere in the QoL for persons with diabetes (Kiadaliri et al., 2013). QoL is also known to be related to physical activity and individuals with diabetes who are physically active tend to have higher QoL scores than those that are inactive (Maharaj and Nuhu, 2015; Çolak et al., 2015).

Little information is documented on PA levels in persons with DM in developing countries. While Adeniyi et al. (2010) found that patients with diabetes were more physically active at moderate levels, Oguntibeju et al. (2012) indicated that a large proportion of these individuals had a low PA level. Significant positive relationship has been shown between PA levels and QoL of individuals with diabetes (Çolak et al., 2015) indicating that PA has a positive impact on QoL. A third study also found that the majority of the participants (with diabetes) had low PA levels (54.7%) (Shiriyedeve et al., 2019).

There is a need for data to be provided about the pattern of physical activity, sleep duration and health-related quality of life among persons with diabetes given the dearth of such information particularly in resource-poor settings. Specifically, the assessment of sleep disorders is important in persons with DM (American Diabetes Association, 2018) as sleep anomaly acting alone or in conjunction with any of the other complications and co-morbidities can severely interfere with physical and psychosocial well-being thereby compromising overall quality of life. Information obtained from this study can serve to stimulate stakeholders in diabetes management to make recommendations and, possibly, create or modify approaches to enhance quality of care as it pertains to non-pharmacological management of diabetes. This study was, thus, undertaken to explore the pattern of physical activity, self-reported sleep and quality of life among individuals with diabetes mellitus in the two major Kano State-owned health facilities in Kano, Northwestern Nigeria.

Study setting and design

A cross sectional survey was used to gather information from persons with diabetes mellitus attending diabetes clinics for follow up at Murtala Muhammad Specialist Hospital (bed capacity of 826) and Muhammad Abdullahi Wase Teaching Hospital (320-bedded) in Kano, Northwestern Nigeria. These are two large referral secondary health facilities (located in the metropolis) owned by the Kano State Government which serve the people of Kano State including those from neighboring states (Hadiza, 2018). Convenience sampling technique was used to recruit the participants who were approached and recruited at the diabetes clinics of the two hospitals.

Data collection procedure

Prior to the commencement of the survey, ethical approval was sought and obtained from the Research Ethics Committee of Kano State Ministry of Health. Potential respondents were approached and the purpose and importance of the study was clearly explained. They were also informed that participation in the survey was voluntary and that they were at liberty to withdraw their consent/participation at any time or decline from filling in the questionnaires. The study was conducted between July 2019 and January 2020 and data were collected at the diabetes clinic of the two hospitals.

All questionnaires were translated from English to Hausa by a qualified translator for the purpose of our study. This was to maintain consistency and stability with interviewer administration of the questionnaires in the target language for those not familiar with English. Questionnaires were either self- or interviewer-administered. Interview was conducted by the researchers and trained interviewers. Majority of the participants (n = 76) self-completed the survey.

Assessment of physical activity level

Respondents’ physical activity (PA) levels were assessed using the short form International Physical Activity Questionnaire (IPAQ-SF). Due to its ease of completion relative to the long form, the IPAQ-SF is recommended for surveys where time is limited (Craig et al., 2003). The types of intensity of PA and sitting time accrued by individuals as part of their daily lives are considered for estimating total PA and time spent sitting. The instrument consists of seven items and measures the level of PA (expressed in MET-minutes/week [www.ipaq.ki.se]) across four domains (leisure time PA, domestic and gardening activities, work-related PA, and transport-related PA). In each domain, the duration (in minutes) and frequency (days) of PA including sitting, walking, moderate and vigorous PA are self-reported (Lee et al., 2011).

The IPAQ was shown to be reliable (p = 0.76) and at least as valid as other physical-activity measures (concurrent = 0.58; criterion = 0.30 against the Computer Science Accelerometer) (Craig et al., 2003). It has also been shown to be reliable and valid in obtaining comparable estimates of PA, based on activity undertaken across multiple domains including leisure, domestic, work and transport (Hagströmer et al., 2006). The Hausa IPAQ-SF has acceptable concurrent validity and test-retest reliability for vigorous-intensity activity, walking, sitting and total PA, but demonstrated only fair construct validity for moderate and sitting activities. The Hausa IPAQ-SF can be used for physical activity measurements in Nigeria, but further construct validity testing with objective measures such as an accelerometer is needed (Oyeyemi et al., 2011). To determine levels of PA, the following values are used:

Walking MET = 3.3 × walking minutes × walking days;

Moderate MET = 4.0 × walking minutes × walking days;

Vigorous MET = 8.0 × walking minutes × walking days and

Total Physical Activity MET = sum of Walking + Moderate + Vigorous MET minutes/week scores (www.ipaq.ki.se).

Respondents were asked to report on their PA during a typical week. Specifically, they were asked to report the number of days and duration they were engaged in vigorous, moderate, walking activity, and total PA. They were classified into low, moderate and high PA based on their weekly energy expenditure in accordance with the IPAQ scoring protocol on physical activity categories and cut-off levels (Appendix).

Measurement of quality of life

The short-form SF-36 health questionnaire was used to assess the quality of life (QoL) of the respondents. This is an eight-domain generic health-related QoL measure whose validity and reliability have been shown as an assessment tool used to indicate changes in health status (McDowell, 2009) in a wide range of diseased conditions. The domains or subscales which include Mental Health, Role Emotional, Social Functioning, Vitality, General Health, Bodily Pain [BP], Role Physical and Physical Functioning are used collectively to reflect QoL scores (Klevsgård et al., 2002; McDowell, 2009). Scores range from 0 to 100 (expressed as percentages) with higher scores indicating better health (Maruish et al., 2011). For ease of interpretation, a norm-based scoring, with a mean of 50 and a standard deviation of 10 in the general population, is employed. Two component summary scores (the Physical Component Summary [PCS] and the Mental Component Summary [MCS]) are frequently used and are derived when scores from the domains are summed up (Ware and Ware, 1993). The instrument’s reliability (median reliability coefficient 0.85 for all subscales) been established (McHorney et al., 1994), and its validity was shown in an elderly population, in which it distinguished between those with and without poor health (Lyons et al., 1994). Although no Hausa translated version of the SF-36 health survey has been documented, the shorter version (SF-12) was demonstrated to have satisfactory construct validity, internal consistency, and test-retest reliability in Hausa-speaking patients with chronic LBP (Ibrahim et al., 2020).

Measurement of sleep duration

Information on respondents’ sleep duration (that is, total amount of sleep experienced during the nocturnal sleep episode), was obtained via a single item measure by asking the question “How many hours of sleep did you experience last night?”). Single-item measures are considered reasonable when the construct is concrete and well defined (Konstabel et al., 2017). In addition, they are less expensive to administer and analyse. Although multiple-item measures are preferable from a clinimetric standpoint, a single-item measure could be useful as long as it discriminates between different groups of patients and reflects clinically relevant changes over time (Fava et al., 2012). Respondents reported daily when they slept the previous night and the time they woke up in the morning (Buysse et al., 1989). This nighttime sleep duration was calculated in hours and the mean for the past seven days recorded.

In a study by St-Onge et al. (2016) on sleep duration and quality in cardiometabolic health, inadequate (or short) sleep duration was defined as <7 h per night, and long sleep was described as ≥9 h per night (St-Onge et al., 2016).

Measurement of weight and height

Body weight was measured (at 0.1 kg precision) with a weighing scale (Tanita BWB-800, Arlington Heights, IL), and height (at 0.5 cm scale precision) using a Holtain Harpenden portable stadiometer (Holtain Ltd., Crymych Pembs, UK). These measurements were taken in light clothing without footwear according to standardized protocols (Stewart et al., 2011). Body mass index (BMI) was calculated as weight (kg)/height2 (m2) (Janssen et al., 2002) and was based on standard BMI categories established by the National Heart, Lung, and Blood Institute (1998) as underweight, <18.5; normal weight, 18.5-24.9; overweight, 25.0-29.9, grade 1 obesity, 30.0-34.9, grade 2 obesity, 35.0-39.9, and grade 3 obesity, ≥40.0.

Information about respondents’ age, gender, and duration since diagnosis of diabetes was obtained by way of interviewing the respondents and confirmed using their case notes.

Data analysis

Data were analyzed using descriptive (frequency, mean and percentage) and inferential (Spearman correlation) statistics using the Statistical Package for the Social Sciences version 20 (SPSS Inc, Chicago, Illinois, USA). A probability level of ≤0.05 was considered significant.

Measurement of weight and height

Body weight was measured (at 0.1 kg precision) with a weighing scale (Tanita BWB-800, Arlington Heights, IL), and height (at 0.5 cm scale precision) using a Holtain Harpenden portable stadiometer (Holtain Ltd., Crymych Pembs, UK). These measurements were taken in light clothing without footwear according to standardized protocols (Stewart et al., 2011). Body mass index (BMI) was calculated as weight (kg)/height² (m²) (Janssen et al., 2002)and was based on standard BMI categories established by the National Heart, Lung, and Blood Institute (1998) as underweight, <18.5; normal weight, 18.5-24.9; overweight, 25.0-29.9, grade 1 obesity, 30.0-34.9, grade 2 obesity, 35.0-39.9, and grade 3 obesity, ≥40.0.

Information about respondents’ age, gender, and duration since diagnosis of diabetes was obtained by way of interviewing the respondents and confirmed using their case notes.

Data analysis

Data were analyzed using descriptive (frequency, mean and percentage) and inferential (Spearman correlation) statistics using the Statistical Package for the Social Sciences version 20 (SPSS Inc, Chicago, Illinois, USA). A probability level of ≤0.05 was considered significant.

Socio-demographic characteristics of the respondents

A total of 124 individuals (76 female and 48 male) diagnosed with diabetes mellitus took part in the study.

With a mean age of 54.61, majority of the respondents were within the 40 to 60 years age category and had duration since diagnosis of diabetes of less than 10 years (Table 1). There were more respondents with scores indicating lower than average quality of life in physical component of the SF-36 health survey than those with scores ≥50. This is the case about scores for the mental component of the survey (Table 1). About 60% of them had normal body mass index (Table 1) and over half of them were in the low physical activity category (Figure 1). About half of the respondents reported that they experienced 6 to 8 h per day. However, a greater number (one-third) of them felt that their sleep was adequate (Figure 2).

Correlation between physical activity and clinical variables

No statistically significant correlation was found between physical activity and either of the two clinical variables (sleep duration and QoL). In addition, sleep duration did not significantly correlate with respondents’ quality of life (Table 2).

phosphorylation of PKB is linked to insulin resistance through negative insulin receptor substrate functions (Zick, 2001), reduced phosphatidylinositol 3-kinase (PI3K) activity (Tremblay et al., 2001), and impaired phosphorylation of the PKB substrate AS160 (Karlsson et al., 2005). Short sleep duration has also been shown to be associated with increased levels of inflammation markers (Williams et al., 2007), and sleep disruption may be related to diabetes via the mechanisms of low-grade systemic inflammation and sleep-inducing effects of the inflammatory state (Zizi et al., 2010). Thus, sleep of short duration is associated with higher risk for various health outcomes including diabetes.

The non-significant correlation between PA and sleep duration was found to be due to the majority of the respondents experiencing either short sleep duration or longer than the optimal number of hours for sleep which may exacerbate their condition (diabetes). This might be true about the non-significant correlation between PA and QoL as well as that between sleep duration and QoL. These non-significant correlations might also be attributed to majority of the respondents being within the low PA category.

This study was not without limitations. First, the correlations between PA and socio-demographic variables of the selected population were not explored. Second, the study did not assess the predictors of PA, QoL and sleep among diabetic patients. In addition, our sample size was not based on power calculation. Thus, the results are not a precise reflection of the characteristics of all patients with DM in the population and cannot, therefore, be generalizable. However, we are optimistic that our findings have enriched the literature. Future studies may be conducted to address these issues.

It was concluded that the majority of the individuals with diabetes mellitus in the selected health facilities were not meeting any of the criteria for either moderate or high levels of physical activity with sleep duration in a significant proportion of them being suboptimal for good quality of life and general wellbeing. Emphasis should be made on information about the benefits of regular physical exercise and adequate sleep in diabetes management.

The authors have not declared any conflict of interests.