To explore primary care physicians’ perspectives on possible barriers to the use of insulin.

This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Eight electronic databases were searched (between January 1, 1994 and August 31, 2014) for relevant studies. A search for grey literature and a review of the references in the retrieved studies were also conducted. Studies that focused on healthcare providers’ perspectives on possible barriers to insulin initiation with type 2 diabetic patients were included, as well as articles suggesting solutions for these barriers. Review articles and studies that only considered patients’ perspectives were excluded.

A total of 19 studies met the inclusion criteria and were therefore included in this study: 10 of these studies used qualitative methods, 8 used quantitative methods and 1 used mixed methods. Studies included a range of different health care settings. The findings are reported under four broad categories: The perceptions of primary care physicians about the barriers to initiate insulin therapy for type 2 diabetes patients, how primary care physicians assess patients prior to initiating insulin, professional roles and possible solutions to overcome these barriers. The barriers described were many and covered doctor, patient, system and technological aspects. Interventions that focused on doctor training and support, or IT-based decision support were few, and did not result in significant improvement.

Primary care physicians’ known delay in insulin initiation is multifactorial. Published reports of attempts to find solutions for these barriers were limited in number.

Diabetes is a major world health problem. The most recent report from the International Diabetes Federation estimates that 415 million people have diabetes worldwide (8.8% of the adult population). Of these, 193 million people have undiagnosed diabetes, which allows the disease to progress untreated causing many complications. In 2015, diabetes was responsible for 11.6% of total global adult health expenditure and yet there were 5 million deaths from this disease. In addition, the number of people with diabetes is expected to increase to more than 642 million by 2040[1].

Type 2 diabetes is characterized by defects in both insulin secretion (due to a progressive decline in beta cell functioning), and insulin resistance. It is aggravated by obesity and sedentary lifestyle. Untreated hyperglycemia increases the risk of mortality and morbidity, via a higher risk of macro-vascular and micro-vascular complications[2-4].

With already high numbers of patients and a relatively small number of diabetes specialists worldwide, 90% of patients receive care for their diabetes from primary care physicians (PCPs)[5,6].

Several trials[7,8] have shown that improving glycemic control, via lifestyle modifications and the use of medications, reduces micro-vascular and possibly macro-vascular complications and mortality related to diabetes. However, over time, the progressive nature of beta cell dysfunction results in the inability of oral hypoglycemic agents to control hyperglycemia and achieve HbA1c targets[9]. This gradual loss in beta cell function would indicate that insulin therapy is almost always required at some point to treat diabetic patients[10].

Although most PCPs believe that the initiation of insulin therapy is an essential component in the management of type 2 diabetes, many still consider it to be the “last option” and indicate that their patients are reluctant to accept this therapy. In the seminal Diabetes Attitudes, Wishes, and Needs (DAWN) study, Peyrot et al[11] reported that approximately 50% of healthcare professionals delay insulin initiation until it is “absolutely necessary”.

Similarly, the SOLVE™ (Study of Once Daily Levemir), a multicenter observational study that involved over 17374 patients with type 2 diabetes in 10 countries (Europe, Asia and North America), showed that insulin initiation is generally delayed until an average HbA1c level of approximately 9%[12]. Several other studies across many countries have confirmed that there is significant delay in the initiation of insulin therapy[13-15]. This reluctance to initiate insulin treatment may be related to patient, provider or system factors.

Reported patient-related barriers include a sense of personal failure, a negative impact on social life, injection phobia, myths and misconceptions about the drug, the permanence of the therapy, difficulties in fulfilling responsibilities at home and at work, limited insulin self-management training, inadequate provider explanation about the risks and benefits of the intervention - and concerns over weight gain and hypoglycemia[16-21]. Most of these negative attitudes and perceptions are described in the literature under the term “psychological insulin resistance”. Studies in Various countries have found variable rates of such resistance amongst patients - from 5.9% to nearly 50% of patients[17-25].

System barriers to the use of insulin therapy may include a lack of resources (e.g., staff and materials), a lack of continuity of care, as well as the workload and time constraints of PCPs[26].

As for the providers, part of the problem has been attributed to “therapeutic or clinical inertia; a recognition (by the PCP) of a lack of glycemic control - but a failure to act”[27-29]. These PCPs blame (perceived) patient reluctance, language barriers, their concern for patients’ comorbidities and their own lack of training[26].

A few systematic reviews have been published related to patients’ perspectives on the delay in insulin initiation or to psychological insulin resistance[30-33]. This study focuses on PCP’s perspectives on their barriers to initiating insulin therapy in primary care/general practice. It will also explore the literature for possible solutions.

This literature search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[34]. As this research doesn’t involve experiments on humans or animals, ethical approval is not required.

A variety of scholarly databases (PubMed, EMBASE, Psychinfo, Cinahl, Ovid, Scopus, ProQuest and the Cochrane library) were searched. The search was conducted using a combination of medical subject headings and text words: Insulin, physician, general practitioner, general practice, primary care, health care, professional, provider, psychological, clinical inertia, attitude, delay, refusal, refuse, perception, postpone, worry, initiation, compliance, knowledge, fear, belief and barrier.

A search in Google and Google Scholar engines were then conducted to look for “grey” literature and the bibliographies of the retrieved articles were also reviewed to identify further relevant studies.

Studies that satisfied the following selection criteria were included: Only original articles, quantitative or qualitative methods, descriptive or analytical, written in English and published between January 1, 1994 and August 31, 2014.

This review was restricted to studies that focused on PCP’s perspectives to the barriers they faced regarding the initiation of insulin for type 2 diabetic patients. Articles suggesting solutions for these barriers were also included. Review articles or studies that only explored patients’ perspectives were excluded.

The screening and selection of the studies were done independently by the two authors, and the final number of the studies used for this review was reached by consensus between them. The authors assessed the methods used in the various studies: The Critical Appraisal Skills Program checklist[35] was used for the qualitative studies, the STROBE checklist[36] was used for the observational studies, the Mixed Methods Appraisal Tool[37] was used for the mixed-methods studies and the Centre for Evidence-Based Medicine tool[38] was used for the randomized control trials.

The authors had to be in agreement for an article to be included in the systematic review. Both authors also collaborated to extract the data from each article. These are summarized in table form according to authors’ name, year of publication, aim of the study, sample size, the setting, instruments used, results and conclusion (Table 1).

The initial search process resulted in 9757 studies. After removing any duplicates and adding articles by searching the cross-references and grey literature, this number dropped to 6683 studies. Titles and abstracts of these studies were then screened, and the two authors narrowed the results down to 132 articles for full-text assessment (78 articles were then excluded because they looked only at patients’ perspectives, and a further 32 were excluded because they were review studies).

The remaining 22 articles were critically appraised by the two authors, and three more articles were excluded because of their poor overall quality. Thus, a total of 19 articles were included in the final analysis. Figure 1 is a flow diagram of this process.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Flow diagram of the studies included in this analysis.

To the best of our knowledge, the present study is the first systematic review of published work that has explored primary care physicians’ points of view regarding the noticeable delay in insulin therapy initiation in primary care settings.

This clinical inertia affects approximately one-third of patients with type 2 diabetes, and this failure of PCPs to act, coupled with patients’ prolonged exposure to hyperglycemia, leads to a higher risk of chronic complications and mortality[28].

Our review finds that PCPs report many barriers to the initiation of insulin and these range across doctor, patient and system issues. Others have also pointed to this complexity[26,39,41,51]. Table 2 listed the main barriers reported by PCPs to initiate insulin as found in this review.

Clearly, the knowledge, training and experience of the PCP is an important factor, and must affect patient management in more domains than just insulin initiation[11]. This factor may also overlap with an over-concern (by the PCP) about their patient’s ability to cope with injections or about possible patient side-effects such as weight-gain and hypoglycemia[26,39,42-44].

Less obviously connected to training and experience is the finding that PCPs under-estimate their patient’s feelings of guilt and personal failure when faced with insulin initiation. About 40% of diabetic patients report poor psychological well-being; guilt and self-blame may well be included in this[56].

And PCPs may assess their insulin-requiring diabetic patients in ways different to their specialist colleagues - considering patient adherence and motivation (as well as general health status, HbA1C and blood glucose levels)[46,47]; thus delaying insulin initiation. Yet a concern for this wider assessment seems to miss relevant patient concerns such as the impacts (of insulin use) on their social, marital, occupational or financial situations, as well as causing inconvenience and social stigma. Patients also report lack of understanding of the illness and the role of insulin[48,49]. Perhaps PCPs hear this patient reluctance and assume it is mainly about the use of injections.

Counter to this is the reassuring findings of two studies that suggest patients’ negative attitudes toward insulin are mostly temporary and improve after they begin using insulin[57,58]. Perhaps patients need education and reassurance earlier in their diabetic journey; that insulin will most probably be needed at some stage, despite their best efforts, and that this is part of the natural history of the disease - not a sign of their poor management and certainly not a reason for self-blame.

There have been various opinions and suggestions for improving the timeliness of insulin initiation. These have included better delineation of roles within the primary health care team, better integration of health care services, improved communications within the health care setting, improved education of patients (including in groups), wider use of nursing staff, decision support for PCPs, or harnessing new technologies[52-55]. These are all consistent with the suggested approach captured within the Chronic Care Model[59].

Some of the above ideas have attracted intervention studies, with varying success and/or methodological rigor: Some interventions have targeted the PCP (including how to deal with barriers to insulin initiation) but there are few in the literature that report clinically positive effects. Two have evaluated such interventions and found that PCPs reported “increased confidence” in using insulin therapy[53,54].

But with a definite clinical outcome measure (HbA1C over three years), Dale et al[60] demonstrated a sustained effect from a brief training course on insulin initiation for healthcare professionals. The training involved “presentations, small group work, case studies and practical demonstrations”.

On the other hand, Harris et al[55] (in a randomized controlled trial) could not demonstrate any clinical effect from an intervention that provided external expert consultation support to PCPs by community pharmacists and specialists.

Another controlled trial used a decision-support tool aimed at helping PCPs make decisions about insulin therapy. They found no difference in insulin initiation between control and intervention groups - though physicians were allowed not to consult the decision support tool[61].

Clearly, interventions aimed at the PCP will need further refinement, especially if they are to be scaled up for the future increasing numbers of diabetic patients.

Of course technology may change the discourse about clinical inertia or psychological insulin resistance in other ways. For example prefilled insulin pens could reduce the time needed for patient education and training. This method is associated with increased patient convenience and persistence[62]. Other technical developments can also be expected. Table 3 listed the possible solutions for insulin initiation delay proposed in the literature.

We gratefully acknowledge Dr. Peter Schattner for conducting a final review before official submission of the manuscript.