Primary care physician assistant and advance practice nurses roles: Patient healthcare utilization, unmet need, and satisfaction

Team-Based Care Model Improves Timely Access to Care and Patient Satisfaction in General Cardiology

ABSTRACT

Appointment wait times have increased nationally since 2014, especially in cardiology. At a mid-Atlantic academic medical center, access to care in the general cardiology clinic was below national standards, which can negatively affect patient outcomes and satisfaction. Adopting a team-based care (TBC) model, advanced practice providers (APPs) were added to care teams with general cardiologists to provide timely outpatient management of cardiac conditions. This aimed to increase access to care and, consequently, patient satisfaction. A formative program evaluation using the Agency for Clinical Innovation framework assessed TBC's impact on these outcomes. Access to care and patient satisfaction measures for TBC and nonteam providers were compared with one another and national benchmarks. Nine months after implementation, the average time to new patient appointment for TBC providers was 31 days (47% decrease) and for nonteam providers was 41 days (20% decrease). TBC had a higher percentage of new patient appointments within 14 days than nonteam providers (39% and 20%, respectively). Patient satisfaction improved to the 98th percentile nationally for TBC but decreased to the 71st percentile for nonteam. These findings suggest that a TBC model using APPs can improve access to care and patient satisfaction in the outpatient general cardiology setting.

Utilization of Medicare's chronic care management services by primary care providers

BACKGROUND

Medicare billing codes introduced in 2015 reimburses primary care providers for non-face-to-face, chronic care management (CCM) services rendered by clinical staff.

PURPOSE

The purpose of this manuscript was to describe provider trends in billed CCM services and identify factors associated with CCM utilization.

METHODS

Observational study using Medicare Public Use Files, 2015 to 2018. General, family, geriatric, and internal medicine physicians, nurse practitioners (NPs), and physician assistants (PAs) with billed primary care services were included. Multivariable analyses modeled associations between the CCM services and type of provider, adjusting for year, primary care services, practice, and patient characteristics.

FINDINGS

Among 140,465 physicians and 141,118 NPs/PAs, CCM services increased each year, yet remained underutilized: 2% to 7% of physicians and 0.3% to 1.3% of NPs/PAs billed CCM in 2018. Increases in beneficiaries (p < .0001), percentage of dually enrolled (p = .0134), and primary care services (p < .0001) predicted higher CCM utilization.

DISCUSSION

CCM utilization reflects practice-based efforts to improve patient access to care by enhancing care delivery.

Physician Associates/Assistants in Primary Care: Policy and Value

Since the new century, primary care physician supply has worsened. Analysts predict that health service demand in the United States will grow faster than physician supply. One strategy is the utilization of physician assistants/associates (PAs). Most PAs work full-time, and approximately one quarter are employed in family medicine/general medicine. PAs deliver primary care services in a team-oriented fashion in a wide variety of settings, including private health systems and community health centers. One fifth work in rural and medically underserved areas. Together PAs and nurse practitioners provide approximately one third of the medical services in family medicine, urgent care, and emergency medicine.

Beyond Patient-Provider Relationships: Expanding the Roles and Boundaries of Families during Patient End-of-Life

Role conflict and strain occur when healthcare providers are required to cross boundaries, either voluntarily or involuntarily, to meet the needs of their dying patients. This research is an unobtrusive digital ethnography of a publicly accessible online forum for healthcare providers (N = 242 posts); it explores the boundaries set by families and healthcare providers, and identifies how healthcare providers navigate and which circumstances require them to sometimes cross these professional boundaries. Results indicate that patient-and-family-centered care may not be fully achieved due to the ambiguity in the expected roles played by both families and healthcare providers during patient death and dying. Grounded in data, an expanded model of the therapeutic alliance, which includes the family, is suggested.

Teaming up in primary care: Membership boundaries, interdependence, and coordination

OBJECTIVE

Increased demand for quality primary care and value-based payment has prompted interest in implementing primary care teams. Evidence-based recommendations for implementing teams will be critical to successful PA participation. This study sought to describe how primary care providers (PCPs) define team membership boundaries and coordinate tasks.

METHODS

This mixed-methods study included 28 PCPs from a primary care network. We analyzed survey data using descriptive statistics and interview data using content analysis.

RESULTS

Ninety-six percent of PCPs reported team membership. Team models fell into one of five categories. The predominant coordination mechanism differed by whether coordination was required in a visit or between visits.

CONCLUSIONS

Team-based primary care is a strategy for improving access to quality primary care. Most PCPs define team membership based on within-visit task interdependencies. Our findings suggest that team-based interventions can focus on clarifying team membership, increasing interaction between clinicians, and enhancing the electronic health record to facilitate between-visit coordination.

Psychometric properties of two instruments measuring self-efficacy and outcome expectations of providing inhaler technique education to patients

INTRODUCTION

Both the National Heart, Lung, and Blood Institute (NHLBI) and Global Initiative for Asthma (GINA) asthma practice guidelines recommend that providers routinely check inhaler technique and correct any mistakes that patients may make when using these devices. Providers, however, rarely check inhaler technique during asthma visits. The objectives of this study were to: (1) describe the development of an instrument to measure self-efficacy and outcome expectations regarding inhaler technique patient education, (2) evaluate the internal consistency reliability of the new scales, and (3) provide preliminary evidence of construct validity. Methods: First- and second-year physician assistant (PA) students at two institutions completed an anonymous and voluntary survey evaluating two new instruments, the Teaching Inhalers to Patients: Self-efficacy (TIP-SE) and the Teaching Inhalers to Patients: Outcome Expectations (TIP-OE) scales and sociodemographic characteristics. The data were analyzed using Principal Components Analysis (PCA), Cronbach's α, and multivariable logistic regression. Results: We had usable responses from 146 PA students (71.9% participation rate). The PCA identified one factor for the TIP-SE and TIP-OE, respectively. The internal consistency of the TIP-SE and TIP-OE was α = 0.96 and α = 0.92, respectively. The logistic regression found that second-year PA students who had higher mean TIP-SE scores were significantly more likely to report teaching patients to use inhalers during rotations (OR = 1.8, 95% CI = 1.1, 2.9). There was not a statistically significant relationship between reporting teaching patients to use inhalers during rotations and mean TIP-OE scores. Conclusion: The TIP-SE and TIP-OE show preliminary evidence of reliability and validity.

Supplemental data for this article is available online at https://doi.org/10.1080/02770903.2021.2008428 .

Clinician Care Team Composition and Health Care Utilization

Objective

To test the hypothesis that a greater proportion of physician time on primary care teams are associated with decreased emergency department (ED) visits, hospital admissions, and readmissions, and to determine clinician and care team characteristics associated with greater utilization.

Patients and Methods

We retrospectively analyzed administrative data collected from January 1 to December 31, 2017, of 420 family medicine clinicians (253 physicians, 167 nurse practitioners/physician assistants [NP/PAs]) with patient panels in an integrated health system in 59 Midwestern communities serving rural and urban areas in Minnesota, Wisconsin, and Iowa. These clinicians cared for 419,581 patients through 110 care teams, with varying numbers of physicians and NP/PAs. Primary outcome measures were rates of ED visits, hospitalizations, and readmissions.

Results

The proportion of physician full-time equivalents on the team was unrelated to rates of ED visits (rate ratio [RR] = 0.826; 95% confidence interval [CI], 0.624 to 1.063), hospitalizations (RR = 0.894; 95% CI, 0.746 to 1.072), or readmissions (RR = –0.026; 95% CI, 0.364 to 0.312). In separate multivariable models adjusted for clinician and practice-level characteristics, the rate of ED visits was positively associated with mean panel hierarchical condition category (HCC) score, urban vs rural setting, NP/PA vs physician, and lower years in practice. The rate of inpatient admissions was associated with HCC score, and 30-day hospital readmissions were positively associated with HCC score, lower years in practice, and male clinicians.

Conclusion

Care team physician and NP/PA composition was not independently related to utilization. More complex panels had higher rates of ED visits, hospitalization, and readmissions. Statistically significant differences between physician and NP/PA panels were only evident for ED visits.

Utilization and Costs by Primary Care Provider Type: Are There Differences Among Diabetic Patients of Physicians, Nurse Practitioners, and Physician Assistants?

OBJECTIVE

The objective of this study was to compare health care utilization and costs among diabetes patients with physician, nurse practitioner (NP), or physician assistant (PA) primary care providers (PCPs).

RESEARCH DESIGN AND METHODS

Cohort study using Veterans Affairs (VA) electronic health record data to examine the relationship between PCP type and utilization and costs over 1 year in 368,481 adult, diabetes patients. Relationship between PCP type and utilization and costs in 2013 was examined with extensive adjustment for patient and facility characteristics. Emergency department and outpatient analyses used negative binomial models; hospitalizations used logistic regression. Costs were analyzed using generalized linear models.

RESULTS

PCPs were physicians, NPs, and PAs for 74.9% (n=276,009), 18.2% (n=67,120), and 6.9% (n=25,352) of patients respectively. Patients of NPs and PAs have lower odds of inpatient admission [odds ratio for NP vs. physician 0.90, 95% confidence interval (CI)=0.87-0.93; PA vs. physician 0.92, 95% CI=0.87-0.97], and lower emergency department use (0.67 visits on average for physicians, 95% CI=0.65-0.68; 0.60 for NPs, 95% CI=0.58-0.63; 0.59 for PAs, 95% CI=0.56-0.63). This translates into NPs and PAs having ~$500-$700 less health care costs per patient per year (P<0.0001).

CONCLUSIONS

Expanded use of NPs and PAs in the PCP role for some patients may be associated with notable cost savings. In our cohort, substituting care patterns and creating similar clinical situations in which they practice, NPs and PAs may have reduced costs of care by up to 150-190 million dollars in 2013.

Patient Acceptability of the Use of Advanced Practice Providers in an Outpatient Neurosurgery Clinic

Introduction

Increasing demands for healthcare manpower has necessitated the utilization of advanced practice providers (APPs). The effect of APPs in primary care has been well-characterized but is less studied in surgical subspecialties. The objective of this study is to assess the patient acceptability of APPs in an outpatient neurosurgery setting.

Methods

We conducted a prospective, survey-based study among 78 adult patients in the neurosurgical outpatient clinic. The survey consisted of 10 questions assessing the hypothetical acceptability of care provided by neurosurgeons and APPs. These were compared as pre-specified dyads, with patients blinded to dyad composition. The data were analyzed with Chi-square tests.

Results

Patients preferred to see their neurosurgeon for their first clinic visit even with a longer lag time (29% acceptability difference, p = 0.012). Patients also preferred to see the neurosurgeon for their first postoperative visit (20% difference, p = 0.009). For all visits, patients preferred to see an APP if the clinic visit would be on time, rather than see the surgeon with a significant delay (30% difference, p = 0.0002). If their visit was scheduled with an APP, patients preferred that the neurosurgeon review their treatment plan before they left the clinic (15% difference, p = 0.04). Overall, seeing an APP was acceptable if patients were informed ahead of time (37% difference, p < 0.0001).

Conclusions

Team-based care utilizing APPs is acceptable to patients. Patients had strong preferences for seeing their surgeon for the first neurosurgical clinic visit and first post-operative visit. Patients were satisfied with seeing an APP if they could be seen more expeditiously. Patients also preferred to know ahead of time if they were going to see an APP.

Collaborative practice trends in US physician office visits: an analysis of the National Ambulatory Medical Care Survey (NAMCS), 2007-2016

OBJECTIVE

Practice arrangements in physician offices were characterised by examining the share of visits that involved physician assistants (PAs) and nurse practitioners (NPs). The hypothesis was that collaborative practice (ie, care delivered by a dyad of physician-PA and/or physician-NP) was increasing.

DESIGN

Temporal ecological study.

SETTING

Non-federal physician offices.

PARTICIPANTS

Patient visits to a physician, PA or NP, spanning years 2007-2016.

METHODS

A stratified random sample of visits to office-based physicians was pooled through the National Ambulatory Medical Care Survey public use linkage file. Among 317 674 visits to physicians, PAs or NPs, solo and collaborative practices were described and compared over two timespans of 2007-2011 and 2012-2016. Weighted patient visits were aggregated in bivariate analyses to achieve nationally representative estimates. Survey statistics assessed patient demographic characteristics, reason for visit and visit specialty by provider type.

RESULTS

Within years 2007-2011 and 2012-2016, there were 4.4 billion and 4.1 billion physician office visits (POVs), respectively. Comparing the two timespans, the rate of POVs with a solo PA (0.43% vs 0.21%) or NP (0.31% vs 0.17%) decreased. Rate of POVs with a collaborative physician-PA increased non-significantly. Rate of POVs with a collaborative physician-NP (0.49% vs 0.97%, p<0.01) increased. Overall, collaborative practice, in particular physician-NP, has increased in recent years (p<0.01), while visits handled by a solo PA or NP decreased (p<0.01). In models adjusted for patient age and chronic conditions, the odds of collaborative practice in years 2012-2016 compared with years 2007-2011 was 35% higher (95% CI 1.01 to 1.79). Furthermore, in 2012-2016, NPs provided more independent primary care, and PAs provided more independent care in a non-primary care medical specialty. Preventive visits declined among all providers.

CONCLUSIONS

In non-federal physician offices, collaborative care with a physician-PA or physician-NP appears to be a growing part of office-based healthcare delivery.

Association of Primary Care Team Composition and Clinician Burnout in a Primary Care Practice Network

Objective

To determine the relationship of the emotional exhaustion domain of burnout with care team composition in a Midwestern primary care practice network.

Participants and Methods

We studied 420 family medicine clinicians (253 physicians and 167 nurse practitioners/physician assistants [NP/PAs]) within a large integrated health system throughout 59 Midwestern communities. The observational cross-sectional study utilized a single-question clinician self-assessment of the emotional exhaustion domain of burnout on a scale of 0 (never) to 6 (daily) conducted between March 1 and April 2, 2018, and administrative data collected between January 1, 2017, and December 31, 2017. We used a multivariable linear mixed model for data analysis, adjusted for clinical- and team-level factors, including clinician sex, panel size and complexity, clinician type (physician or NP/PA), clinician full-time equivalent (FTE), total care team panel size, and number of clinicians on the care team.

Results

Among 217 survey respondents (51.7%), the median frequency of the emotional exhaustion domain of burnout was once per week. Adjusted analyses revealed that a greater proportion of physician FTE on the care team was associated with a lower emotional exhaustion domain of burnout among individual clinicians (P=.05). Female clinicians had a higher emotional exhaustion domain of burnout than male clinicians (P=.05). None of the other variables in the model were associated with emotional exhaustion.

Conclusion

Primary care teams containing both physicians and NP/PAs had lower levels of emotional exhaustion with increasing proportion of physician FTE. More work is needed to explore what other variables may be associated with burnout in primary care team-based practices.

An activity analysis of Dutch hospital-based physician assistants and nurse practitioners

BACKGROUND

The physician assistant (PA) and the nurse practitioner (NP) were introduced into The Netherlands in 2001 and 1997 respectively. By the second decade, national policies had accelerated the acceptance and development of these professions. Since 2015, the PA and NP have full practice authority as independent health professionals. The aim of this research was to gain a better understanding of the tasks and responsibilities that are being shifted from Medical Doctors (MD) to PAs and NPs in hospitals. More specifically in what context and visibility are these tasks undertaken by hospital-based PAs and NPs in patient care. This will enable them to communicate their worth to the hospital management.

STUDY DESIGN

A descriptive, non-experimental research method design was used to collect and analyze both quantitative and qualitative data about the type of tasks performed by a PA or NP. Fifteen medical departments across four hospitals participated.

METHODS

The patient scheduling system and hospital information system were probed to identify and characterize a wide variety of clinical tasks. The array of tasks was further verified by 108 interviews. All tasks were divided into direct and indirect patient care. Once the tasks were cataloged, then MDs and hospital managers graded the PA- or NP-performed tasks and assessed their contributions to the hospital management system.

FINDINGS

In total, 2883 tasks were assessed. Overall, PAs and NPs performed a wide variety of clinical and administrative tasks, which differed across hospitals and medical specialties. Data from interviews and the hospital management systems revealed that over a third of the tasks were not properly registered or attributed to the PA or NP. After correction, it was found that the NP and PA spent more than two thirds of their working time on direct patient care.

CONCLUSIONS

NPs and PAs performed a wide variety of clinical tasks, and the consistency of these tasks differed per medical specialty. Despite the fact that a large part of the tasks was not visible due to incorrect administration, the interviews with MDs and managers revealed that the use of an NP or PA was considered to have an added value at the quality of care as well to the production for hospital-based medical care in The Netherlands.

What Can the Giant Do? Defining the Path to Unsupervised Primary Care Practice by Competence, Not Time

In this issue of Academic Medicine, Dewan and Norcini examine the significant variability of time-in-training between patient care "giants"-the physicians, nurse practitioners, and physician assistants who practice primary care-and they call for further studies to determine optimal training duration and eventual scope of practice. They ask, what is the minimum education and training required to practice primary care, or "how tall is the shortest giant?" In this Invited Commentary, the authors reframe the question from identifying the minimum length of training required, to identifying desired patient care outcomes. Primary care is not a uniform entity. It ranges from complex elderly chronically ill patients, to twentysomething millennials with acute problems, to pregnant women, to families, and everything in between. The authors argue that training should be fit for purpose and produce high-quality outcomes for patients. Competence should be defined by these outcomes. Drawing parallels with Major League Baseball, the authors note that time to competence development will be variable for different training programs depending on purpose, and also variable for people within those programs, even with shared purpose. While time is a tool for competence attainment, it should not be the metric by which readiness for practice is measured.

Resource Utilization in Non-Academic Emergency Departments with Advanced Practice Providers

INTRODUCTION

Advanced practice providers (APP), including physicians' assistants and nurse practitioners, have been increasingly incorporated into emergency department (ED) staffing over the past decade. There is scant literature examining resource utilization and the cost benefit of having APPs in the ED. The objectives of this study were to compare resource utilization in EDs that use APPs in their staffing model with those that do not and to estimate costs associated with the utilized resources.

METHODS

In this five-year retrospective secondary data analysis of the Emergency Department Benchmarking Alliance (EDBA), we compared resource utilization rates in EDs with and without APPs in non-academic EDs. Primary outcomes were hospital admission and use of computed tomography (CT), radiography, ultrasound, and magnetic resonance imaging (MRI). Costs were estimated using the 2014 physician fee schedule and inpatient payments from the Centers for Medicare and Medicaid Services. We measured outcomes as rates per 100 visits. Data were analyzed using a mixed linear model with repeated measures, adjusted for annual volume, patient acuity, and attending hours. We used the adjusted net difference to project utilization costs between the two groups per 1000 visits.

RESULTS

Of the 1054 EDs included in this study, 79% employed APPs. Relative to EDs without APPs, EDs staffing APPs had higher resource utilization rates (use per 100 visits): 3.0 more admissions (95% confidence interval [CI], 2.0-4.1), 1.7 more CTs (95% CI, 0.2-3.1), 4.5 more radiographs (95% CI, 2.2-6.9), and 1.0 more ultrasound (95% CI, 0.3-1.7) but comparable MRI use 0.1 (95% CI, -0.2-0.3). Projected costs of these differences varied among the resource utilized. Compared to EDs without APPs, EDs with APPs were estimated to have 30.4 more admissions per 1000 visits, which could accrue $414,717 in utilization costs.

CONCLUSION

EDs staffing APPs were associated with modest increases in resource utilization as measured by admissions and imaging studies.

Intermediate Diabetes Outcomes in Patients Managed by Physicians, Nurse Practitioners, or Physician Assistants: A Cohort Study

BACKGROUND

Primary care provided by nurse practitioners (NPs) and physician assistants (PAs) has been proposed as a solution to expected workforce shortages.

OBJECTIVE

To examine potential differences in intermediate diabetes outcomes among patients of physician, NP, and PA primary care providers (PCPs).

DESIGN

Cohort study using data from the U.S. Department of Veterans Affairs (VA) electronic health record.

SETTING

568 VA primary care facilities.

PATIENTS

368 481 adult patients with diabetes treated pharmaceutically.

MEASUREMENTS

The relationship between the profession of the PCP (the provider the patient visited most often in 2012) and both continuous and dichotomous control of hemoglobin A1c (HbA1c), systolic blood pressure (SBP), and low-density lipoprotein cholesterol (LDL-C) was examined on the basis of the mean of measurements in 2013. Inverse probability of PCP type was used to balance cohort characteristics. Hierarchical linear mixed models and logistic regression models were used to analyze continuous and dichotomous outcomes, respectively.

RESULTS

The PCPs were physicians (n = 3487), NPs (n = 1445), and PAs (n = 443) for 74.9%, 18.2%, and 6.9% of patients, respectively. The difference in HbA1c values compared with physicians was -0.05% (95% CI, -0.07% to -0.02%) for NPs and 0.01% (CI, -0.02% to 0.04%) for PAs. For SBP, the difference was -0.08 mm Hg (CI, -0.34 to 0.18 mm Hg) for NPs and 0.02 mm Hg (CI, -0.42 to 0.38 mm Hg) for PAs. For LDL-C, the difference was 0.01 mmol/L (CI, 0.00 to 0.03 mmol/L) (0.57 mg/dL [CI, 0.03 to 1.11 mg/dL]) for NPs and 0.03 mmol/L (CI, 0.01 to 0.05 mmol/L) (1.08 mg/dL [CI, 0.25 to 1.91 mg/dL]) for PAs. None of these differences were clinically significant.

LIMITATION

Most VA patients are men who receive treatment in a staff-model health care system.

CONCLUSION

No clinically significant variation was found among the 3 PCP types with regard to diabetes outcomes, suggesting that similar chronic illness outcomes may be achieved by physicians, NPs, and PAs.

PRIMARY FUNDING SOURCE

VA Health Services Research and Development.

Interpersonal continuity of primary care of veterans with diabetes: a cohort study using electronic health record data

Background

Continuity of care is a cornerstone of primary care and is important for patients with chronic diseases such as diabetes. The study objective was to examine patient, provider and contextual factors associated with interpersonal continuity of care (ICoC) among Veteran’s Health Administration (VHA) primary care patients with diabetes.

Methods

This patient-level cohort study (N = 656,368) used electronic health record data of adult, pharmaceutically treated patients (96.5% male) with diabetes at national VHA primary care clinics in 2012 and 2013. Each patient was assigned a “home” VHA facility as the primary care clinic most frequently visited, and a primary care provider (PCP) within that home clinic who was most often seen. Patient demographic, medical and social complexity variables, provider type, and clinic contextual variables were utilized. We examined the association of ICoC, measured as maintaining the same PCP across both years, with all variables simultaneously using logistic regression fit with generalized estimating equations.

Results

Among VHA patients with diabetes, 22.3% switched providers between 2012 and 2013. Twelve patient, two provider and two contextual factors were associated with ICoC. Patient characteristics associated with disruptions in ICoC included demographic factors, medical complexity, and social challenges (example: homeless at any time during the year OR = 0.79, CI = 0.75–0.83). However, disruption in ICoC was most likely experienced by patients whose providers left the clinic (OR = 0.09, CI = 0.07–0.11). One contextual factor impacting ICoC included NP regulation (most restrictive NP regulation (OR = 0.79 CI = 0.69–0.97; reference least restrictive regulation).

Conclusions

ICoC is an important mechanism for the delivery of quality primary care to patients with diabetes. By identifying patient, provider, and contextual factors that impact ICoC, this project can inform the development of interventions to improve continuity of chronic illness care.

Comparative evaluation of patients' and physicians' satisfaction with interferon beta-1b therapy

Background

Due to the preventive nature of disease-modifying therapies for multiple sclerosis, treatment success particularly depends on adherence to therapeutic regimens and patients’ perception of treatment efficacy. The latter is strongly influenced by the confidence in the involved health care professionals and the relationship to the treating physician.

Methods

In this report, we considered physicians’ and patients’ evaluation of satisfaction with interferon beta-1b treatment efficacy for assessing the congruence in ratings. Data were queried in a study conducted between 2009 and 2013.

Results

After 6 months of therapy, > 80 % of the patients and physicians (N = 445) showed high degrees of satisfaction regarding interferon beta-1b treatment, with only few physicians and patients (≤2.0 %) rating “not satisfied”. The proportion of patients rating with the same category as their physicians was similar after 6 months (47 % congruence) and at the 24 months/study end visit (49 %). Discrepancies between ratings were observed with respect to study end: for patients with premature study end, more patients and physicians rated being not satisfied with the therapy, accompanied by a considerably lower congruence of 33 % compared to 54 % for patients receiving the therapy for at least 2 years and completing the study regularly.

Conclusions

Regular communication between physicians and patients about their perception of therapy might improve alignment of treatment evaluation and could result in increased therapy persistence. In addition, patients’ willingness to perform a long-term therapy − even in the absence of disease symptoms − might be promoted by repeated exchange between health care providers and patients with regard to realistic treatment expectations.

Trial registration

ClinicalTrials.gov NCT00902135 (registered May 13, 2009).

Electronic supplementary material

The online version of this article (doi:10.1186/s12883-016-0705-1) contains supplementary material, which is available to authorized users.