Leveraging Registered Dietitian Nutritionists and Registered Nurses in Medication Management to Reduce Therapeutic Inertia Free

A literature search was conducted to identify published studies that met defined inclusion criteria. To be included in the review, studies needed to be reported in English, published between 1 January 2000 and 31 December 2019, and describe interventions that used RDNs or RNs to deliver pharmacological therapy using protocols to initiate or adjust medications for diabetes, dyslipidemia, or hypertension in adults ≥18 years of age. Additional criteria included outpatient/ambulatory care, study duration ≥6 months, and a completion rate ≥80%. Study designs included randomized controlled trials (RCTs), retrospective and prospective cohort studies, and time series (pre-/post- comparisons). The electronic databases searched included PubMed, the Cochrane Central Register of Controlled Trials, Ovid, and the Cumulative Index to Nursing and Allied Health Literature. Each database was independently searched for articles using the following medical subject headings: diabetes mellitus, dyslipidemia, hypertension, drug therapy, dietitians or dieticians, and nurses. Boolean “AND” was used to combine the drug therapy, disease state terms, and health care professionals.

The literature search identified 4,477 citations. An additional 40 studies of interest were identified using an ancestry approach. Titles and abstracts were then screened to determine whether the articles were relevant. After applying inclusion criteria and removing duplicates, 210 full-text articles were requested for further review to assess for eligibility (Figure 1).

Ultimately, 20 studies met the inclusion criteria, including 12 RCTs, one retrospective cohort (i.e., chart review) study, six prospective cohort studies (including two with pre-/post- comparisons), and one time series (pre-/post- comparison) (Table 1) (13–32). In all, the studies represent 7,280 participants with a median study duration of 12 months (range 6–25 months). Thirteen of the studies included a lifestyle behavior change component in addition to use of medication therapy protocols.

Table 2 (13–35) summarizes the following characteristics for the 20 studies identified: type of health care professional delivering the intervention, chronic condition addressed, and presence or absence of detailed information on the protocol or treatment algorithm used. Fifteen studies used RNs to deliver the interventions, two used RDNs, and three used both RDNs and RNs. Nine studies described the medication protocol. Among the studies included in our systematic review, typically the RDN- and RN-led medication therapy arm was compared with usual care, whereby usual care signified care delivered by a “primary care provider” (PCP), “general practitioner,” “physician,” “primary care physician,” internist,” or “diabetologist” (14,15,18–21,23–25,29,31,32).

Interventions led by RDNs and RNs led to improvements in blood pressure, lipids, and glycemic control. In studies with a comparison group available (n = 17), outcomes were as good as or better than with usual care and often led to a higher percentage of participants achieving treatment goals (Table 1). Compared with usual care, improvements in the RDN/RN interventions were significantly better in five intervention studies for A1C (15,17,18,21,32), six studies for blood pressure (21,23–25,28,29), and four studies for lipids (13,17,21,23). Of the three studies without a usual care comparison group (22,27,30), improvements were seen for the RDN-/RN-led intervention in target outcomes.

Of the 20 total studies, 15 were led by RNs. Of these, 12 focused on blood pressure, and all 12 saw an improvement in blood pressure with the RN intervention. Of those with a comparison group (n = 11), six studies found a statistically significant improvement over usual care (21,23–25,28,29) and five were noninferior (17,19,20,31,32), with the final pre-/post- study achieving statistically significant reductions in both systolic and diastolic blood pressure (22). Eight RN-led studies focused on diabetes (A1C), and all but one had a comparison group. Of these, four had significant improvements over the comparison arm (17,18,21,32), and three were noninferior (16,19,20). The final study found significant improvement in reducing A1C via the intervention, with 45% achieving an A1C <7% (22). Finally, eight studies assessed the effectiveness of RN-led interventions on dyslipidemia, with seven of the interventions having a comparison group. Of these, four studies found a significant improvement for lipids (17,20,21,23) and three were noninferior (19,26,32). The prospective cohort study by McLoughney et al. (22) also significantly improved total cholesterol and triglycerides, with 91% achieving lipid targets.

Two study designs looked at the effectiveness of RDNs using a treatment protocol. In a telemedicine RCT focused on type 2 diabetes, there was a statistically significant improvement in composite optimal care measures (A1C, blood pressure, not using tobacco, and taking a statin and aspirin as appropriate) in the intervention group compared with the control group (14). In a time series (pre-/post- comparison) study, total cholesterol and triglyceride levels decreased significantly, and these levels were maintained at 1 year (30).

In the three studies jointly implemented by both RDNs and RNs, there were improvements in clinical outcomes. In an RCT in type 2 diabetes comparing case management to traditional primary care in a Medicaid population, A1C decreased in both groups (by an average 1.88%) and was sustained throughout the 2-year study. The reduction in A1C was statistically greater in the intervention group at each time point (15). In a primary prevention cardiovascular retrospective cohort study (13), more participants than nonparticipants achieved an LDL cholesterol level <100 mg/dL and quit using tobacco. More encounters were associated with greater improvements in total cholesterol and LDL cholesterol than with fewer encounters. Finally, the study by Senior et al. (27) found statistically significant improvements in A1C, blood pressure, and LDL cholesterol (27).

The results of the 20 articles included in this review (Table 1), in the context of available literature, make a compelling case that collaborative team-based interventions that use medication therapy protocols followed by RDNs and RNs improve clinical outcomes. Of the 20 studies, all demonstrated improvements in A1C, blood pressure, or lipids with 13 studies providing a lifestyle behavior change component in addition to medication protocols. Twelve of the 20 primary studies (57%) were RCTs, which included a total of 4,315 participants.

Guidelines from the AHA, ACC, and ADA consistently recommend a collaborative, multidisciplinary approach to meet patient needs (1–3). Under this approach, such professionals, with appropriate training and supervision, can follow an approved treatment algorithm to make medication changes for patients without consulting a PCP (36).

Current ADA guidelines recommend a change of medication therapy if A1C targets are not met after 3 months, although the reported times are generally much longer. A systematic review by Khunti et al. (37) found that the median time to treatment intensification was >1 year (range 0.3 to >7 years) when an A1C was above target in most studies reviewed. This failure to intensify treatment in a timely manner comes at a significant cost, according to Ali et al. (38). They estimated that, in a patient population of just over 13 million individuals with type 2 diabetes and an A1C >9%, delaying medical therapy intensification was associated with a loss of 13,390 life-years and a cost of $7.3 billion.

This review highlights the ability to overcome therapeutic inertia with the use of expanded care teams. Furler et al. (18) found that 70% of patients were started on insulin in the intervention group compared with 22% in the control group. Zgibor et al. (32) also found that intervention participants were much more likely to have their medication intensified compared with usual care. In another study focused on managing hypertension (25), 97% of the intervention group had one or more changes in drug therapy at 6 months compared with 43% in the usual care group. The intervention group also demonstrated greater medication adherence.

There are a multitude of reasons for clinical and therapeutic inertia. Competing demands faced by physicians during medical encounters present a major barrier to the provision of preventive services. These demands include, but are not limited to, addressing patient-related factors (e.g., fear of hypoglycemia, weight gain, injections, and myalgia), clinician-related factors (e.g., an ever-increasing number of medications; frequent changes to clinical practice guidelines, leading to knowledge gaps; and differing comfort levels with offering new advances in treatment), and health system–related factors (e.g., medication costs, limited access to care, and lack of timely access to relevant clinical data) (39,40). Given that social determinants of health account for 50–60% of health outcomes, it is also of increasing importance to address these factors during clinic visits (41). Compounding these competing demands are often severe time limitations resulting from high patient loads; most PCPs have no more than 10–15 minutes to spend per patient visit (37). Finally, the United States is currently experiencing a significant shortage of PCPs, as demand for physicians continues to grow faster than supply (42).

RDNs and RNs may be able to mitigate some of these issues by expanding their role to include medication management in addition to lifestyle behavior change counseling. This comprehensive approach to care could be integrated with the PCP visit (i.e., as a shared appointment) or by serving as an extension of care between usual clinic visits to the PCP. As an extension of primary care, RDNs and RNs are likely able to provide more frequent visits (or as needed) compared with usual care by a PCP, which typically occurs every 6–12 months (14,15,21–23). Frequency of intervention visits versus control visits often was not reported in these studies; yet, it is likely an important factor in achieving therapeutic outcomes. More encounters can lead to greater improvement in outcomes (13) and sustained improvement (30), while patients may feel more satisfied with the care they receive (16,19,20).

While cost was not addressed in most studies reviewed, Houweling et al. (20) did find lower health care costs and higher patient satisfaction with the diabetes nurse specialist intervention compared with usual care provided by an internist.

Most studies included in this review had several strengths, including a thorough description of the inclusion and exclusion criteria for the patient population selected, integration of the interventions within existing clinic or care delivery processes, and, in many cases, use of electronic health records as part of the intervention and for data surveillance.

This systematic review has some limitations. Efforts were made to capture all available published studies related to the aim of the review. However, it should be emphasized that selection and publication bias might be present and that both are inherent to the research modality chosen (i.e., a systematic review). Also, many articles reviewed for possible inclusion mentioned medication management but did not specify that protocols or treatment algorithms were used. As a result, the articles were excluded.

Most of the RCTs were not blinded, so interventionists knew which patients were in which group. Although most of the studies had specific follow-up time points built into the intervention design, most did not describe the average number of visits needed to achieve success.

As with all research and program evaluations, there is the potential that results were influenced by program participants being more motivated to participate and make lifestyle behavior changes compared with nonparticipants.

Although nearly all the studies adequately described the most important characteristics of the patients and interventions, most did not adequately describe the process used to develop and approve protocols within their institutions, the number of health professionals used to deliver the interventions, the educational background of the interventionists, or any additional training provided to the RDNs and RNs who assumed this role.

Organizations who are looking to pursue models that use medication therapy protocols need to check state licensure laws for RDNs and RNs. If the law does not address this issue, it becomes a matter of local facility discretion. Facilities can decide whether RDNs and RNs are able to initiate or titrate medications within protocols that have been approved by their own governing body.

Training for RDNs and RNs to pursue this type of work is an important consideration. In this review, the studies ranged from no formal training to intensive training. For example, in the study by Senior et al. (27), the RDNs and RNs delivering the intervention attended a 5-day training course before the study began covering topics such as disease management, pharmacotherapy, and the use of treatment algorithms, ongoing monthly all-day training sessions, and biweekly telehealth sessions and had ongoing peer mentoring by an advanced-practice nurse.

Collaboration with PCPs and other care team members is a crucial element of success to establish trust for this expanded role. Senior et al. (27) emphasized the importance of establishing trust and strong communication to overcome any concerns of patients and referring providers related to health professionals assuming nontraditional roles. For these professionals, access to a supportive and expert central team was crucial to building clinic staff confidence and reassuring physicians that the care of their patients was appropriately supervised.

Future research related to these types of interventions should more clearly specify whether a protocol was used and how it was developed and provide sufficient detail outlining the scope of the protocol. This will help other facilities and professionals learn how this type of work can best be implemented. Additionally, future studies would benefit from including cost as a specific outcome of interest, as this was addressed very minimally by the studies available to date.

This systematic review provides evidence that RDN- and RN-led medication management using physician-approved protocols or treatment algorithms can lead to clinically significant improvements in diabetes, dyslipidemia, and hypertension management. A collaborative, multidisciplinary approach involving traditionally nonprescribing health care professionals in the care of patients managed by PCPs can help to ensure that these patients receive both timely adjustment of their therapeutic regimens and essential coaching for lifestyle behavior change.

No potential conflicts of interest relevant to this article were reported.

G.B. participated in the conception of the study and drafting of the manuscript. J.H. researched data and participated in drafting of the manuscript. T.B.-L. researched data and edited the manuscript. A.S. analyzed and interpreted the data. J.B. participated in the conception of the study and edited the manuscript. G.B. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.