I need help with an article analysis.

1. What is the hypotheses?

2. What are the independent and dependent variables and types of data for variables

3. Sample and sampling method

4. How data was collected

The full article is below for reference.

Sterzi, D.,Auziere, S.,Glah, D., & Markert Jensen, M. (2017). Perceptions of general practitioners on initiation and intensification of type 2 diabetes injectable therapies. A quantitative study in the United Kingdom.Primary Care Diabetes,11(3), 241-247.

Abstract

Most diabetes care is done by general practitioners (GPs) in the UK. This study aimed to determine GPs' comfort level in initiating and intensifying injectable therapies, identifying any associated barriers, and assessing reasons for referral to specialists. This web-interview included 128 general practitioners (GPs) experienced in type 2 diabetes (T2D) management, as well as 57 specialists and 30 nurses who were studied for secondary objectives. GPs felt more comfortable initiating the 1st injectable therapy - typically the glucagon-like peptide-1 receptor agonists (GLP-1 RA) - than the 2nd. The main barriers to initiating injectables were related to the complexity of injectable therapies and the lack of comfort with complex patient profiles, namely patients with difficultly achieving glycaemic control or those with significant comorbidities who GPs would rather refer to specialists. The main attributes that would increase their comfort level with initiation of injectables are improved glycaemic control, weight control and low risk of hypoglycaemia. An injectable therapy with these attributes could help to overcome barriers to initiating injectable therapies among GPs other healthcare professionals in primary care.

1.Introduction

Mostdiabetesmanagement is undertaken byprimary careproviders: only 20% of people with diabetes ever see anendocrinologist[1]. However, the increasing complexity of diabetes management drives primary care providers (PCPs) to refer patients to specialists, or to maintain patients at suboptimalglycaemiclevels. Manypatients with type 2 diabetes(T2D) are not intensified in a timely manner: UK general practice data showed that three years after basal insulin initiation, only 29% of patients haveglycated haemoglobin(HbA1c) levels 7%[2], and 60% of T2D patients are maintained on basal insulin and oralanti-diabetic(OAD) treatments, despite evidence of poorglycaemic control[3].

This study aimed to determine GPs' comfort level in initiating and intensifying injectable therapies. This included the identification of any potential barriers they may face with initiation or intensification, as well as drivers for referral to specialists.

2.Methods

2.1.Study design

This was a national, computer-assisted web interview to assess the perceptions of GPs on injectable T2D therapy. The leading objective was to determine GPs' comfort levels and identify any potential barriers in initiating and intensifying injectable therapy, as well as what drives referral to specialists. Specialists and nurses' perceptions of injectable initiation and intensification inprimary carewere also explored as a secondary objective assigned to the study.

The study was approved by a European Ethics committee. No patients' data were collected; therefore no informed consent was necessary. The study was conducted in the UK from November 2014 to January 2015.

2.2.Selection of participants

Participants were contacted through a medical physicians' panel (SERMO). They were selected based on: number of years in practice, their role in the management of T2D, time spent in primary care, proportion of time spent with the patient during a consultation, number of T2D patients seen/treated in a month, and number of T2D patients treated with insulin in a month (Table 1). These selection criteria aimed to capture the view of physicians who are experienced in the management of T2D. Physicians meeting selection criteria underwent a 25min online interview.

2.3.Information collected

Participants answered questions that covered the following topics: comfort level in treating T2D patients and initiating/intensifying injectable therapy, reasons for initiating injectable therapies, attributes that would increase the comfort level, frequency and reasons for referrals to specialists for initiating/intensifying injectable therapies, specialty to which the patient is referred, attributes of an "ideal" injectable therapy and drivers for the referral to specialists. Initiation of a 1st injectable therapy was defined as initiating aglucagon-like peptide-1 receptor agonist(GLP-1 RA) or basal insulin with patients currently treated with OAD. Intensification with a 2nd injectable was defined as adding a basal insulin with patients currently treated with aGLP-1RA (OADs), or adding either a GLP-1 RA or a bolus insulin in patients currently treated with basal insulin (OADs).

2.4.Statistical analysis

Descriptive statistics (means, standard deviations, median, min and max values) were provided for continuous variables such as age, and number of years of practice or number of T2D patients seen per month. Categorical variables such as gender, dichotomous questions or open-ended questions were summarised as frequencies and percentages when applicable. Confidence intervals were also provided to ensure precision of range of values reported.

Missing values were not replaced. Comparisons used z-test, Chi²-test or analysis of variance, as appropriate. P-value for all tests was set at a significance level of 0.05.

In assessing the drivers to refer a T2D patient to a specialist for initiation of an injectable therapy, GPs were asked to assess a set of attributes that could influence their decision. Data were analysed using a Maximum Difference Scaling technique[4]. This technique allows to best discriminate and evaluate a large number of attributes to reveal which amongst them would be the most influential and which would be the least influential. The exercise consisted of seven scenarios, each scenario showing a group of three attributes appearing on a screen at once. The attributes shown for each scenario were determined by an experimental design. In each scenario, GPs were asked to select the attribute they believed was the "most important/influential" in their decision to refer a T2D patient to a specialist and then the "least important/influential". Separate questions were asked for the 1st and 2nd injectable. Results are expressed in "winning percentages"[4]: the modelled percentage of time an attribute is chosen as the "most important/influential" vs. other attributes based on the experimental design.

3.Results

3.1.Characteristics of participants and current practice

After screening, 128 GPs were selected to participate. Their mean age (SD) was 45 years (8.4), with 17.26 (7.2) years in practice. Details on the current practice of participants are summarised inTable 1.

Specialists involved in the study managed a higher number of patients treated with injectable therapies, and initiated injectable therapy more frequently than GPs. In a typical month, less than one quarter of T2D patients requiring an injectable therapy were referred to specialists. GPs and specialists agreed onHbA1clevels at which patients needed intensification. This is observed amongst specific patient subgroups, such as obese patients, patients with renal impairment or significantcomorbidities, patients older than 65 years old and patients in need of 3rd party assistance (Fig. 1). No significance difference was observed between GPs' and specialists' perceptions.

3.2.Comfort level of initiating or intensifying with injectable therapies among GPs

Table 2summarises the comfort level of GPs in initiating an injectable therapy. GPs felt more comfortable initiating the 1st injectable than the 2nd injectable. However, the majority of GPs referred patients to specialists for the initiation of an injectable therapy. While GPs mainly referred toendocrinologists, the study shows thatprimary carenurses would refer todiabetesspecialist or endocrinologists equally.

Although few GPs felt very comfortable with initiating injectable therapy, they indicated that they would like to increase level of initiation of injectable therapies. This willingness was more marked for the 1st injectable, especially forGLP-1RA, for which around half of GPs stated to be very comfortable to initiate (42%; CI:8.6%) compared to basal insulin, for which around a third of them expressed high comfort level (36%; CI8.3%). In contrast, only 15% of GPs would like to intensify therapy with bolus insulin. Nevertheless, this willingness tends to increase to 20% with GLP-1 RA added on to basal insulin; primary care nurses' perception of their comfort level was similar to those of GPs.

3.3.Reasons for initiating injectables and attributes of an "ideal" injectable therapy

GPs mentioned three key reasons for initiating injectable therapy. The most important was related to the improvement ofglycaemiccontrol, stated by more than two-third of them (69%), then confidence in initiating treatment (40%) and thirdly having more experience with injectable therapies (34%).

As for attributes for an "ideal" injectable therapy, more than half of the GPs (58%) mentioned reliable glycaemic reduction. In addition, 40% also stated an ideal injectable therapy with no side effects (e.g. hypoglycaemia, nausea, diarrhoea) would be ideal. For approximately one third of the GPs (35%), a weight neutral orweight reducinginjectable therapy as well as simplicity of use for patients (29%) would also be relevant. These ideal attributes are shared by primary care nurses who also added no renal or cardiaccontraindication(47%; P<0.05 nurses vs. GPs). As a whole, more than half of GPs (56%) would feel very comfortable initiating an injectable therapy bearing these ideal attributes.

3.4.Attributes that would further increase the comfort level with initiating injectable therapy

Fig. 2shows the attributes which would have the greatest impact on GPs' comfort level with initiation or intensification. It is worth noting that treatment related attributes appear to be of key importance for the GPs. The foremost attribute that would further increase their comfort level isglycaemic controlimprovement. This attribute is mentioned by 73% of GPs for the 2nd injectable vs. 67% when referring to the 1st injectable. Around 60% mentioned weight control and low risk of hypoglycaemia.

3.5.Drivers to specialist referral for the initiation of an injectable therapy

Fig. 3depicts the drivers for referral to specialists to initiate the 1st or 2nd injectable. The attributes were ranked according to their relative ability to influence treatment referrals[4]. The most influential drivers were mainly patient-related: patients experiencing hypoglycaemia, at high risk of hypoglycaemia, with significant cardiovascular comorbidities, with renal impairment, disagreeing with the recommended treatment, and the need for complicated dose adjustment.

3.6.Specialists' point of view on primary care initiation and intensification of T2D patients with injectable therapies

According to specialists, 62% of the patients initiated on injectable therapy were specifically referred from primary care for this initiation, amongst which 29% were initiated with GLP-1 RA and 26% with basal insulin (as 1st injectable); 21% were initiated with basal insulin+GLP-1 RA and 19% basal+bolus insulin.

Reading through these figures lead to the conclusion that specialists and GPs are in accordance with one another.

Specialists felt very comfortable with GPs initiating the 1st injectable basal insulin (61%) and GLP-1 RA (56%) and the 2nd injectable to a lesser extent; basal+bolus (39%) and basal+GLP-1 RA (42%). Specialists thought the best way to increase GPs' comfort level in initiating injectable therapy is with more training and education.

As a whole, the majority of specialists agreed that patients who are overweight (77%11) or obese (49%13), patients 65 years old (67%12), as well as patients over 65 years old (58%13), could be initiated or intensified equally well in primary care.

Specialists specified the attributes that would increase GPs' ability to initiate more injectable treatments as: clear treatment protocol/guidelines (58%3), ease of use of treatment (47%13) and low risk of hypoglycaemia (44%13). This differed from the GPs' specification of attributes which identified the improvement of glycaemic control and weight neutral, or weight reducing, amongst the top three attributes.

4.Discussion

In the UK, approximately 17% of T2D patients treated with insulin maintain HbA_1Clevels <7%[5], leaving 83% who need intensification. A large proportion of T2D patients (60%) are maintained on basal insulin for three years, despite not achieving the recommended target[3]. This is the first nationwide study of GPs practice in T2D management related to the initiation and intensification of injectable treatments. In this study, nearly three quarters of GPs sought advice from specialists for initiating an injectable, and only 24% of patients were referred for injectable therapy in a given month, which suggests there is clinical inertia in initiating or intensifying patients who require further treatment with injectable therapies. Comfort level is significantly lower for treatment intensification, i.e. initiation of 2nd injectable vs. 1st injectable. Regarding the 1st injectable, this level of comfort is higher forGLP-1RA than for basal insulin, suggesting that treatment complexity is a barrier for PCPs.

This study shows a willingness to increase injectable initiation or intensification if these therapies had the attributes of an "ideal" therapy as described by GPs (Appendix A). These included reliableglycaemicreduction, no side effects (e.g. hypoglycaemia) and weight neutrality or weight loss. Such effective injectable therapies may simplify an initiation inprimary care, mitigating patient related drivers for referral, as well as empowering primary care providers to improveglycaemic controlfor their patients. For the 2nd injectable, basal insulin intensification with a GLP-1 RA has been shown to improve glycaemic control, and may encourage weight loss and provide potential advantages in terms of reducing hypoglycaemia risk[6].

Similarly, this study suggests the factors affecting the comfort level of GPs in initiating injectable treatments in T2D patients may be overcome with these "ideal" attributes as well. This supports a previous study which analysed the barriers to initiatinginsulin therapyin T2D patients in primary health care centres[7]. Patients are referred to specialists when they experience or are at risk of hypoglycaemia, or have significantcomorbidities. T2D patients who need injectable therapies often have frequent comorbidities, and guidelines do not propose explicit recommendations on how to initiate injectable therapies for these specific patient groups. Therefore, there is a need to establish treatment guidelines for more complicated patient subgroups, for example patients experiencing hypoglycaemia, at risk of hypoglycaemia or with co-morbidities renal impairment orcardiovascular diseases. This was also the opinion of specialists in this study who suggested education of PCPs on insulin initiation and easy-to-follow guidelines as focal points that may lead to more initiations in primary care. The perceived complexity may also be a factor for the increased time taken to initiate insulin therapy[8]. This iterates that appropriate education and guidance could further support GPs and potentially reduce the time and lower comfort level with insulin initiation in primary care.

As specialists in this study feel certain patients could be initiated and/or intensified equally well with an injectable therapy in primary care, an integrated care approach between GPs and specialists fordiabetesmanagement could be considered.

Although this study provides some important insights regarding injectable therapy in primary care, there are some limitations of this research. Firstly, in order to obtain meaningful results, the selection criteria meant that respondents were required to have some experience with the management of T2D (Table 1selection criteria). Results are therefore unlikely to represent the views of less experienced practitioners. Secondly, the study relies on self-reporting and is therefore subject to some inherent bias[9],[10],[11].

5.Conclusion

Despite the moderate level of comfort among GPs in initiating injectable therapies shown in the study, there is a significant willingness to initiate more. This willingness is echoed in the specialists' perceptions. Nevertheless, GPs are aware of the challenges they may face with some more complex patients, who they will be more likely to refer to specialists.

Together with an integrated care approach, education and improved guidelines on specific types of patients, injectable therapies that provides reliableglycaemic controlwith a low risk of hypoglycaemia and weight neutrality or weight loss would increase GPs' comfort level to initiate or intensify with these therapy options inprimary care.