Introduction

It is now estimated that between 340 and 536 million adults (aged 20–79 years) worldwide have diabetes, with the disease accounting for an estimated 14.5% of global all-cause mortality in 2015 [1]. However, over the past decade the prevalence of diabetes has risen most rapidly in low- and middle-income countries (LMICs) [2], where it is now estimated that around 75% of people with diabetes live [1]. Low- and middle-income countries also generally lack the resources with which to adequately deal with this growing diabetes epidemic. For example, in 2015 it is estimated that just 19% of global health expenditure on diabetes occurred in LMICs [1].

Poorly managed diabetes results in hyperglycaemia and eventually serious microvascular and macrovascular complications, which can lead to cardiovascular events, kidney failure, blindness, nerve damage and peripheral arterial disease, ultimately resulting in disability and/or mortality [3]. However, tightly controlling blood glucose levels through the effective use of medication and the management of key lifestyle factors, such as diet and exercise, reduces the risk of serious complications developing and progressing [4, 5]. Given the chronic nature of the condition, it is therefore vital that individuals develop the knowledge and skills necessary to effectively manage their condition on a day-to-day basis away from professional healthcare facilities, and thereby prevent or delay the development of complications [6]. Diabetes self-management education (DSME) is therefore seen as “a critical element of care for all people with diabetes and those at risk of developing the disease [6].” However, in LMICs DSME is rarely available or is of poor quality [7], and unlike in many high-income countries [6] context specific guidelines rarely exist for clinical staff to utilise.

A number of meta-analyses of randomised controlled trials (RCTs) have provided support for DSME, demonstrating that compared to routine treatment DSME programmes typically lead to small but significant improvements in blood glucose control [8–11]. Both group-based and individual-based DSME programmes appear effective [8–11], but group-based approaches are typically cheaper, and offer the advantage of allowing patients to meet and discuss their issues together. Unsurprisingly, very few RCTs have evaluated DSME programmes in LMICs (but see [12–15]), and in LMICs people with diabetes typically lack the required skills and knowledge to effectively self-manage their condition [16, 17]. For example, in the Nigerian health system there has been a long-term systematic failure to adequately control blood glucose levels in most diabetes patients [18]. Furthermore, DSME is also only available on a limited basis, health workers often lack adequate knowledge about how to effectively manage patients with diabetes, and the majority of health care facilities do not employ protocol-driven care of patients [19]. Consequently, education of patients and health workers is seen as key to improving this dire situation [18].

Therefore, we developed an intensive, structured and systematic group-based DSME programme that could be delivered by nurses or doctors to people with either type 1 or type 2 diabetes, and was tailored to the Nigerian context. We hypothesised that our intensive group-based DSME would be more effective at helping diabetes patients successfully manage their blood glucose levels compared to the existing unstructured, non-systematic and ad hoc group-based DSME available within a tertiary hospital. This was therefore a more pragmatic and useful comparison than to no DSME.

Methods

For the trial protocol see S1 Text Trial Protocol. The trial was reported according to CONSORT standards: see S2 Text CONSORT Checklist and S1 Table CONSORT Extension for Abstracts Checklist.

Results

A total of 228 individuals were screened and 48 were found to be ineligible (Fig 1). All eligible participants were invited to join the trial. Sixty-two declined to participate, but the remaining 118 eligible individuals provided written informed consent and were randomised equally between treatment groups following collection of baseline data (Fig 1). A total of 14 (11.9%) participants, 6 (10.2%) from the intensive education group and 8 (13.6%) from the conventional education group, were lost to follow-up (Fig 1) after all became uncontactable during the trial, and therefore did not provide outcome data. Outcome data were available for all remaining participants who completed the study: 53 (90%) in the intensive education group and 51 (86%) in the conventional education group (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. CONSORT recruitment and trial participant flow.

https://doi.org/10.1371/journal.pone.0168835.g001

Most participants had type 2 diabetes (85.6%), and were taking oral hypoglycaemic drugs (84.7%) rather than insulin or insulin plus oral hypoglycaemic drugs (14.4%), reflecting the distribution of type 1 and type 2 diabetes patients in the trial population (Table 1). Participants had typically been diagnosed as having diabetes for a number of years (mean 6.5 years), and most were middle-aged (IQR: 48–58.8), clinically overweight (79.7% had a BMI ≥25), and 70.3% had secondary or tertiary education (Table 1). There was some indication of potentially important imbalances in sex, and to a lesser extent education and type of diabetes, but baseline characteristics appeared otherwise well balanced (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline demographic and clinical characteristics by treatment group and overall.

https://doi.org/10.1371/journal.pone.0168835.t001

The primary analysis indicated that participants in the intensive education group had significantly lower HbA_1c levels at six-month follow-up compared to participants in the conventional education group, with a mean estimated difference of -1.8 (95% CI: -2.4 to -1.2) in HbA_1c percentage points (Table 2). This result appeared robust, with no substantive difference found in the results from the adjusted analysis and the analyses using multiply imputed missing data (Table 2 & S3 Text).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Multiple regression results for HbA_1c outcomes at six-month follow-up for participants in the intensive and conventional education groups.

https://doi.org/10.1371/journal.pone.0168835.t002

Discussion

This study demonstrates that an intensive, structured and systematic group-based DSME programme can help patients manage their blood glucose levels more effectively than conventional, ad hoc group-based DSME within a LMIC healthcare setting. Specifically, we found robust results demonstrating that patients in the intensive education group had HbA_1c (%) values at six-month follow-up that were on average -1.8 (95% CI: -2.4 to -1.2) percentage points lower than patients in the conventional education group. This highly statistically significant difference is clinically very meaningful. For example, the UK Prospective Diabetes Study showed that the risk of diabetes complications drops by 35% for every percentage point decrease in HbA_1c [23].

To the best of our knowledge this is the first RCT of a DSME intervention in Nigeria, although a five-year non-randomised study in another Nigerian tertiary hospital compared treatment involving longer doctor-patient consultations and participation in ‘health education forums’ to routine treatment, and provided some evidence of improved blood glucose control and reduced occurrence of morbidity [24]. However, a small number of RCTs evaluating DSME interventions for people with diabetes have been conducted in other LMICs, including Costa Rica, China and Thailand, and have also demonstrated clinically significant improvements in biomedical outcomes [12, 14, 15]. Conversely, an RCT of a DSME programme in under-served communities in South Africa produced no positive results relating to biomedical outcomes, but a process evaluation indicated serious problems with the running of the programme and with patient attendance [13]. Trials of similar diabetes (and CVD) educational guidelines as used here, along with clinical guidelines, are also on-going in Swaziland, Pakistan and China by the COMDIS-HSD group, and the guides, training modules and tools are all freely available online for adaptation in other LMICs (http://comdis-hsd.leeds.ac.uk/resources/technical-guidelines/).

Diabetes self-management education programmes are seen as vital and fundamental tools for enabling patients to effectively manage their diabetes in high-income countries [6]. Therefore, the results presented here, and those from similar trials in LMICs discussed above, indicate that (when properly functioning) DSME programmes can lead to clinically important improvements in patients’ management of their diabetes in LMIC settings. This is an important finding given that the burden from the growing global epidemic of diabetes is felt overwhelmingly in LMICs such as Nigeria [1]. Clearly there appears to be a small but growing base of evidence supporting the roll-out of such programmes in LMIC healthcare settings. Specifically in Nigeria there appears a need for such programmes within general hospitals, as well as both public and private primary care centres, and following the experience in this study the National Council of Health plans to adopt the educational guidelines and roll them out nationally.

This study was conducted in a busy tertiary hospital in Nigeria and pragmatically compared the existing conventional DSME available to patients (representing a best-case scenario for diabetes patients in Nigeria), which involved ad hoc, unstructured and non-systematic didactic provision of information, with a structured, systematic and intensive DSME programme. This programme was based on structured guideline counselling, and the guidelines were designed to be suitable for nurses, educators or doctors in more peripheral hospitals and health centres, and should therefore be replicable in similar settings. Generally, diabetes care in Nigeria is almost solely provided by medical doctors who educate their patients based on their individual experiences and understanding of diabetes care, which can be poor [25, 26]. There are also less than 30,000 doctors in Nigeria today [27], and the approximate average doctor:patient ratio of 1:53,333 indicates that their time should not be spent on educational programmes when such roles may be taken up by nurses or specialist educators. Such a situation is typical for LMICs. Therefore, a nurse- or specialist educator-led approach has the potential to reduce the burden of care for doctors, while being replicable in the primary and secondary levels of care in LMIC health settings where doctors are scarce and their time spread thinly. Clearly though, a major barrier (probably the biggest) to implementing effective DSME programmes in Nigeria and other LMICs is the issue of cost and resources. For example, whilst Nigerian nurses and dieticians overwhelmingly support the introduction of DSME programmes, believing them to be of benefit for their patients, only around one-third believe their establishments would be currently capable of implementing them [28].

The study had a number of limitations. First, the initial six intensive education sessions were led by doctors, when ideally all sessions would have been led by nurses to maximise the generalisability of the results to LMIC settings where nurses or educators would be the most obvious group to run such programmes. However, the education sessions were designed to be simple to run, and we encountered nothing to suggest that nurses, who successfully ran the latter six sessions, would not be able to successfully run the entire programme. Second, the trial only had a six-month follow-up period, and it is not clear how sustainable patient adherence and improvements in biomedical outcomes would be over longer timescales. Third, the mechanisms by which the programme achieved its effects were not explicitly investigated. These two limitations are also shared by the other RCTs of DSME programmes run in LMICs discussed previously [12, 14, 15], where follow-up periods ranged from three to six months, and the mechanisms by which these interventions worked were not explicitly investigated (although the South African trial, which did not produce positive results, did conduct a process evaluation [13]).

Four, it is unclear how generalisable the results are because the patient population and setting of the study do not closely represent the typical Nigerian diabetes patient and health care context. Only 2.5% of participants had no education and 40% had tertiary education, whilst the estimated rate of illiteracy in Nigeria is 40.4% in individuals aged 15 or more [29]. Although the educational materials used during the trial were simplified and relied largely on pictorial representations to drive home key messages, participants with higher levels of education may have benefitted more from the written, and indeed oral, provision of information. It is therefore unclear to what extent the atypically high level of education among participants may have been an enabling factor in the success of the trial. For example, studies have demonstrated positive relationships between the education levels of Nigerian diabetes patients’ and their knowledge about diabetes and its management, and the appropriate use of insulin [16, 17]. The trial was also conducted in just one public teaching hospital, and teaching hospitals are generally better equipped and staffed than either public general hospitals or most private hospitals, as they are directly funded by the federal government. Finally, given the unblinded nature of the study, common to many complex interventions, there is a possibility that trial effects, such as the Hawthorne effect, may have biased the estimated treatment effect of the intervention [30].

In conclusion, this study indicates that an intensive, structured and systematic guideline-based DSME intervention can improve biomedical outcomes in diabetes patients compared to unstructured, non-systematic and ad hoc DSME over relatively short timescales, a finding supported by a small number of similar RCTs in LMIC settings. Combining this structured education, led by nurses or specialist educators, together with quality clinical case management has the potential to improve diabetes management in LMICs, and thereby reduce the frequency of complications and their associated and catastrophic human and financial costs. Future studies of DSME programmes in LMICs should explore the following two sets of key questions. First, how feasible, cost-effective and generalisable are DSME programmes in LMICs? Second, how sustainable are the effects of such programmes in LMIC settings, and by what mechanisms do they generate their effects?

Supporting Information

Acknowledgments

We would like to thank the staff of the Department of Internal Medicine at the University of Calabar Teaching Hospital for supporting various aspects of this clinical trial, and all the patients who participated. We would also like to thank Dr Claire Blackwell for her very useful and comprehensive feedback on an earlier version of the manuscript that helped improve it.