Table 1.

Summary of the articles included in this study

Ref.	Design	Study aims	Sample and setting	Tools and outcome measures	Results and conclusions
Hayat et al[39] (2010)	Qualitative	To explore barriers and myths regarding initiating insulin therapy in poorly controlled type 2 diabetic patients in primary care centres in Hyderabad District, Pakistan	6 to 20 medical officers per each focus group and 12 medical officers for semi-structured interviews Conducted in primary care centres in the Hyderabad District, Pakistan	6 focus group discussions 12 semi-structured individual interviews	Patients barriers: Mistaken beliefs about insulin; fear of needles and excessive belief in traditional healers Doctors’ barriers: Skill and knowledge deficiencies, language barriers and fear of hypoglycaemia and obesity risks System’s barriers: High workload, insufficient consultation time, lack of continuity of care and financial barriers
Manski-Nankervis et al[51] (2014)	Qualitative	To study the effect of communications and relationships between health care professionals in general practice toward the issue of insulin initiation and to clarify how multidisciplinary teams work in practice	21 GPs, practice nurses, diabetes nurse educators and specialist physicians Conducted in Melbourne, Victoria, Australia	Semi-structured interviews face-to-face or via the telephone	Barriers for initiating insulin aren’t solely due to medical or training issues; communication and relationships among health care professionals also have a strong influence 4 themes identified from the study: Uncertain roles, unreliable competency, relationships and communication between healthcare workers and the development of trust and respect
Grant et al[46] (2007)	Cross-sectional	To study physicians’ considerations and preference when selecting medications for patients with type 2 diabetes	886 academic generalists and specialists Conducted in the United States	Questionnaire-based survey	The major considerations for academic generalists when staring insulin are the patient’s HbA1c level, adherence and motivation, health assessment and glucose level patterns Major barriers to beginning insulin for generalists are patient derived (patients’ fear and preferences), while the specialists didn’t specify any major barriers
Yoshioka et al[48] (2013)	Cross-sectional	To identify differences in the perceptions of patients and their physicians regarding insulin initiation (using data from the DAWN Japan study)	148 patients with type 2 diabetes and 68 physicians Conducted in Japan	Questionnaire-based survey	There are perceptions gaps regarding insulin initiation between patients and physicians, especially regarding its social impact Physicians tend to overestimate the barrier of the injections being painful and patients’ fear and to underestimate its social impact
Peyrot et al[11] (2005)	Cross-sectional	To study patients’ and healthcare professionals’ (physicians and nurses) attitudes toward insulin therapy and its correlation with delaying insulin therapy initiation	2061 type 2 diabetic patients and 1109 providers (physicians and nurses) Conducted in 13 countries in Asia, Australia, Europe and North America	Structured interviews conducted face-to-face or over the telephone	Most healthcare professionals (50%-55%) delay insulin initiation until it is absolutely required Delay in insulin initiation is significantly less likely when providers consider that their patients are adherent to appointments and medication regimens Delay of starting oral hypoglycaemic drugs is the strongest correlation with insulin therapy initiation
Lee et al[47] (2014)	Qualitative	To determine how healthcare professionals assess their patients when initiating insulin therapy for type 2 diabetic patients	41 health care professionals (physicians, nurses, diabetic educators and pharmacists) Conducted in Malaysia	4 focus group discussions and 10 individual interviews	Healthcare professionals’ assessment of diabetic patients when considering insulin initiation are based on their perceptions rather than objective evaluation of patients’ backgrounds, knowledge and abilities
Ratanawongsa et al[43] (2012)	Cross-sectional	To explore primary care physicians’ perceptions about barriers of initiating insulin for patients with type 2 diabetes	83 primary care physicians Conducted in United States (Indiana, New jersey and California)	Structured interviews that contained open-ended questions	Participants reported that at least 10% of their patients would reject to start insulin 64% of the clinicians believed that their patients’ reluctance was the cause of delaying insulin initiation; 43% believed it was due to their patients’ poor self-management skills 97% of physicians thought that fear of the injections was the cause of their patients’ resistance to starting insulin
Patel et al[40] (2012)	Qualitative	To identify healthcare professionals’ perspectives on delaying insulin initiation for type 2 diabetic patients in a multi-ethnic setting	14 healthcare professionals (general practitioners, specialists and nurses) Conducted in the United Kingdom	Semi-structured, face-to-face, interviews	Barriers for initiating insulin therapy for South Asian diabetic patients could be over-accentuated by the presence of language barrier and the lack of patients’ understanding about the disease and its therapy South Asian patients seem to be more likely to be negatively influenced by observations and experiences about insulin treatment within their community
Lee et al[41] (2012)	Qualitative	To explore healthcare professionals’ opinions on barriers of initiating insulin therapy in Malaysian multi-ethnic patients with type 2 diabetes	38 healthcare professionals (general practitioners, family medicine specialists, medical officers, policy makers, diabetes educators and endocrinologists) Conducted in Malaysia	Focus group discussions and semi-structured interviews	Patients’ barriers: Patients’ fear and misconceptions about insulin, lack of knowledge and self-efficacy Healthcare professionals’ barriers: Negative attitude toward insulin; lack of training, motivation and confidence System barriers: Lack of continuity of care, shortage of resources and language barriers
Lee et al[52] (2012)	Qualitative	To explore the strategies suggested by healthcare professionals to improve insulin initiation in the Malaysian dual-sector (public-private) healthcare system	38 healthcare professionals (general practitioners, family medicine specialists, medical officers, policy makers, diabetes educators and endocrinologists) Conducted in Malaysia	Focus group discussions and semi-structured interviews	Developing an integrated system for patients’ referral from the private sector to the public health sector for insulin initiation services Involving nongovernmental organisations, the media and pharmaceutical industry in supporting insulin initiation Establishing multidisciplinary teams
Lakkis et al[44] (2013)	Cross-sectional	To investigate family physicians’ attitudes towards insulin therapy in type 2 diabetic patients in Middle Eastern Arab countries	122 family physicians Conducted in Middle Eastern Arab countries	Online questionnaire-based survey	73.6% of family physicians chosen to delay insulin initiation until it is absolutely necessary 64% of family physicians reported hesitancy to start insulin mostly due to apparent patient reluctance
					Referral to endocrinologists for initiating insulin therapy was correlated with a lack of experience and concerns about risks, mainly with elderly patients
Nakar et al[49] (2007)	Case-control study	To study the barriers that delay a shift to insulin from the perspectives of type 2 diabetic patients and family physicians	Study group: 92 patients who needed insulin Control group: 101 patients who had begun insulin 157 family physicians Conducted in Israel	Telephone interviews with participating patients Written questionnaire obtained from the physicians	A clear perception gap regarding the insulin initiation barriers between patients and physicians Physicians exaggerate the importance of patients’ physical fear of pain associated with injections and blood tests, while patient barriers seem to be related more to their concept of the illness
Furler et al[50] (2011)	Qualitative	To describe barriers and facilitators to insulin initiation in general practice	14 healthcare professionals (general practitioners and diabetes nurse educators) and 12 type 2 diabetic patients Conducted in Australia	Semi-structured interviews	Insulin initiation could be influenced by the way patients and healthcare professionals Interact There was a disagreement and uncertainty about the healthcare workers’ role in initiating insulin
Hayes et al[45] (2008)	Cross-sectional	To explore primary care physicians’ attitudes toward insulin initiation for type 2 diabetic patients and to determine the areas where there is a clear lack of consensus between them	505 primary care physicians Conducted in the United States	Online questionnaire-based survey	Highest consensus was attitudes related to risk and benefits of insulin therapy, positive experiences of diabetic patients on insulin and patient worries about insulin initiation Clear lack of consensus was seen in attitudes related to the metabolic effects of insulin, the necessity for insulin therapy, the duration needed for training and the fear of hypoglycaemia risk especially in elderly patients
Tan et al[42] (2011)	Qualitative	To explore the barriers to insulin initiation for diabetic patients managed in primary care polyclinics in Singapore	18 healthcare professionals (physicians and nurses) and 11 type 2 diabetic patients Conducted in Singapore	Focus group discussions	Patient barriers to insulin initiation were denial the need for insulin therapy, perception of social stigma, inconvenience, worries of needles pain, fear from side effects and complications Physician attitude and experience with insulin therapy were also a possible barrier
Haque et al[26] (2005)	Qualitative	To explore the barriers to initiating insulin therapy for type 2 diabetic patients in public-sector primary health care centres in Cape Town, South Africa	46 medical officers working at community health centres Conducted in Cape Town, South Africa	5 focus group discussions and 10 individual semi-structured interviews	Physician barriers: Lack of knowledge and experience, language barriers and exaggerated fear of hypoglycaemia Patient barriers: False beliefs about insulin, poor compliance, lack of understanding of the disease, belief in traditional herbs, fear of injections and poor socioeconomic status System barriers: Time limitations, lack of continuity of care and financial restraints
Sunaert et al[53] (2014)	Qualitative	Related to a program supporting the initiation of insulin therapy in primary care in Belgium, this study determined factors influencing the general practitioners to be involved in insulin therapy initiation and explored factors relevant for future program development	9 general practitioners for focus group discussions 20 general practitioners for individual interviews 10 type 2 diabetic patients for individual interview Conducted in Belgium	Focus group discussions Individual semi-structured interviews	General practitioners whom engaged in insulin initiation program differ from those not engaged in: Attitude, subjective norm and perceived behavioural control regarding insulin initiation Factors to consider include: Job boundaries between the diabetes nurse educators and general practitioners, job boundaries between general practitioners and specialists and protocol adherence
Burden et al[54] (2007)	Mixed	To evaluate the "Insulin For Life" training course for general practitioners and practice nurses in the Heart of Birmingham Teaching Primary Care Trust by exploring the attitudes of the patients, nurses and GPs toward initiating insulin therapy	39 type 2 diabetic patients using a questionnaire 3 to 6 mo after starting insulin 17 general practitioners and practice nurse surveyed using a questionnaire after course completion 37 GPs and practice nurses attended focus group discussions Conducted in Birmingham, United Kingdom	Questionnaire-based survey Focus group discussion	Post-course completion: Type 2 diabetic patients reported that starting insulin in general practice is acceptable and were confident and self-management Most general practitioners and practice nurses were confident about initiating insulin
Harris et al[55] (2013)	Randomised control trial	To determine the effectiveness of an insulin initiation strategy utilising a diabetes specialist and community retail pharmacy support to increase family physician insulin-prescribing rates	73 family physicians in the intervention group were provided with diabetes specialists/educators consultation support for 12 mo and community retail pharmacist support 78 family physicians in the control group had usual care Conducted at 15 sites across Canada	Primary outcome was insulin-prescribing rate (IPR) per physician defined as the number of insulin starts of insulin-eligible patients during the 12-mo period	No significant differences were found between the two groups: Mean IPR of 2.28 compared to 2.29 for Intervention group physicians and the control group physicians, respectively. And an estimated adjusted RR (95%CI) of 0.99 (0.80 to 1.24); P = 0.96 An insulin initiation support program utilising support from diabetes specialists, diabetic educators and community retail pharmacists to improve insulin prescribing in family practice was unsuccessful

DAWN: Diabetes Attitudes, Wishes, and Needs