Audit of Referral Patterns for Hernia Repair – Are General Practitioners Aware of the Changing Face of Herniorrhaphy?

INTRODUCTION

The objective of this study was to examine referral patterns from general practitioners for groin hernia surgery and to assess their knowledge of services available to their patients.

PATIENTS AND METHODS

An anonymous postal questionnaire was sent to 120 general practitioners (GPs) in the South East Wales region who routinely refer patients for inguinal hernia surgery to the Royal Gwent Healthcare NHS Trust.

RESULTS

A total of 86 questionnaire replies were returned. There was variation in referral patterns between the GPs with the majority (84%) referring their patients for groin hernia repair to either a general surgeon or as an open referral. Only 14% referred directly to a hernia specialist and none regularly referred to a laparoscopic surgeon.

CONCLUSIONS

Referral patterns for inguinal hernia surgery do not reflect services provided in secondary care. Further education is required so that a patient''s care can be optimised.     

The fourteen-day rule and colorectal cancer

OBJECTIVE: To examine the effect of the fourteen-day rule on the colorectal service of a district general hospital. METHODS: Prospective audit of all patients referred by general practitioners to the colorectal service of a district general hospital serving a population of approximately 300,000 people. The main outcome measures were: (i) mean interval in days from referral to first clinic appointment; (ii) first clinic appointment to diagnosis; and (iii) overall interval from referral to diagnosis. RESULTS: There was a change in the referral pattern with greater numbers of 'fourteen-day rule' and urgent referrals than expected (P < 0.001). The mean time interval from referral to diagnosis was reduced (P < 0.01). This was due to a reduction in the wait for a first clinic appointment (P < 0.01). The wait between first appointment and diagnosis was unchanged (P < 0.05). Waiting times for patients referred as 'routine' or whose GPs did not specify a priority also improved. CONCLUSIONS: The 'fourteen-day rule' with respect to colorectal cancer has reduced waiting times for a first appointment to see a specialist. Further improvements will require additional resources to reduce the delay for investigations. The effect on long-term survival remains to be seen.     

Inter general practice variability in use of referral guidelines for colorectal cancer

OBJECTIVE: The Two-Week Wait (TWW) referral system for suspected colorectal cancers has a low yield. To examine this, we assessed the referral pattern of general practices within four primary care trusts and looked at the variability of yield of colorectal cancer amongst all TWW referrals and assessed the reasons for variability. METHOD: A prospectively collected database of all colorectal cancers was examined for new cases diagnosed in the 12 months from April 1st 2004. Patients were cross-referenced via general practitioner (GP) codes to identify the referral origin. Reasons for the variability in referral patterns from each general practice were assessed in relation to TWW referrals, population demographics and through postal questionnaire of GPs. RESULTS: A total of 175 patients diagnosed with colorectal cancer were referred from 49 general practices. Whilst there was a positive correlation between the number of TWW referrals and colorectal cancer per 1000-practice population (P = 0.001; Spearman correlation coefficient r(s=0.447,) two-tailed), there was a big discrepancy between referrals and cancer diagnosed in many general practices. Twenty-six general practices (53%) had no colorectal cancer diagnosed via the TWW route and these practices had significantly lower utilization of the TWW referral pathway. In the postal survey, 22% of GPs were unaware of TWW clinics or colorectal cancer referral guidelines and only 8% of GPs knew the number of referral criteria. CONCLUSION: This study demonstrates wide variability within primary care, in the appropriate use of colorectal cancer referral guidelines. General practices should be targeted for education.     

Nurse-led direct access endoscopy clinics: the future?

BACKGROUND: Many studies have shown that rectal bleeding is a good indicator of underlying colorectal pathology, and that ost of the lesions in patients presenting with rectal bleeding lie in the left side of the colon [1, 5, 9, 12, 23, 26]. The recent acceptance of the nurse-practitioner by the National Health Service may allow the use of nurse-endoscopists to develop throughout the United Kingdom. This study aimed to audit a unique nurse-led direct-access nurse-endoscopy service with regard to its efficacy and cost effectiveness, and to monitor patient satisfaction and direct referrals from the primary health sector. METHODS: A nurse-led open-access flexible sigmoidoscopy (OAFS) service for patients reporting fresh rectal bleeding was established at our center in February 1996. A prospective audit of sigmoidoscopic findings and a retrospective analysis of referral patterns from local general practitioners were conducted. A questionnaire survey of both patient and general practitioner satisfaction also was conducted at the same time. RESULTS: Since February 1996, 706 patients have been referred to our service. Rectal bleeding was by far the most common cause for referral, representing the dominant symptom in 92% of the referrals received. Although 99% of the patients underwent a complete sigmoidoscopic examination, 16% of these examinations were limited because of several factors combined. A cause for bleeding was identified in 91% of the patients, with 24% of them experiencing subsequent significant pathology. Of the patients surveyed, 99% were satisfied with the service provided. The results also show nurse-led OAFS to be a more effective use of financial resources, costing $90 less per patient than general practitioner referrals sent to a consultant for further action. CONCLUSIONS: Rectal bleeding is a good indicator of underlying colorectal disease. Most of the significant lesions presenting with this symptom are found in the left side of the colon. A nurse-led OAFS is safe, effective, and acceptable to patients. It also is more cost effective than a consultant-led service.     

Orthopaedic GP Fellowship: does it work?

BACKGROUND: General practitioners (GPs) see a significant number of musculoskeletal problems in their daily caseload. However, orthopaedic training often forms a relatively small part of their undergraduate and postgraduate training. METHODS: A training fellowship for GPs was set up in Warrington to improve management of patients with common orthopaedic complaints in the primary care setting, and to facilitate more appropriate referrals to orthopaedic surgeons. Following the fellowship, GP referral patterns were examined. RESULTS: It was found that the GP fellows were managing many conditions more appropriately, either conservatively, or with skills learnt during the fellowship. There was an increase in the number of referred cases being listed for surgery indicating a more appropriate referral pattern to hospital. CONCLUSIONS: The Orthopaedic GP Fellowship has improved patient management in primary care and helped GPS better identify those patients who need to be referred for a specialist orthopaedic opinion.     

Beurteilung des Verfahrens Hüftendoprothetik durch deutsche Hausärzte

To assess the general knowledge and individual views of general practitioners (GPs) on total hip arthroplasty (THA), we performed a questionnaire-based survey involving 200 GPs in Germany.Ninety-four GPs returned the questionnaire. They had treated a mean of 37.7+/-38.6 patients (range 0-300) with THA. Their general knowledge can be estimated as good. They assessed the potential for complications in association with THA as higher than that reported in the literature. If the GPs had been exposed to more complications in their practices, they estimated a lower patient satisfaction rate 10 years after surgery and were less satisfied with the indications given by their operating colleagues.A broad spectrum of individual responses indicates the need to improve information transfer between orthopaedic surgeons and referring GPs.     

The diagnostic value of digital rectal examination in primary care for palpable rectal tumour

Background Accurate digital rectal examination (DRE) enables the early diagnosis of palpable rectal tumour. We aimed at evaluating the diagnostic value of DRE performed by general practitioners (GPs), with respect to detecting the presence of a palpable rectal tumour. Method All patients diagnosed to have a palpable rectal tumour via a 14‐day cancer referral system between May and December 2006 were identified from the colorectal database. Patients referred by GPs during the same period as having a palpable rectal tumour were also identified by reviewing the 14‐day cancer referrals. Sensitivity, specificity, positive and negative predictive value of a DRE in primary care were calculated by using these data. Results Between May and December 2006, 1069 patients were referred to the University Hospital of North Staffordshire to the 14‐day urgent colorectal cancer referral service. Of these, 108 patients were referred as having a ‘palpable rectal tumour’. Only 32 of the 108 were found to have a rectal lesion on examination in the hospital. Ten tumours were missed by GPs’ DREs. Conclusion Digital rectal examination in primary care for palpable rectal tumour has a sensitivity of 0.762, specificity of 0.917, positive predictive value of 0.296 and negative predictive value of 0.988. It is an inaccurate procedure and a poor predictor for palpable rectal tumour.     

Variation in general practitioner referral of patients with arterial disease

AIM: To compare patterns of referral for arterial disease in two areas in the UK. METHODS: A postal questionnaire was used to survey general practitioner (GP) referral patterns for arterial disease. Questionnaires were sent to West Berkshire GPs in 1993 and 2000 and to Llantrisant GPs in 2000. RESULTS: A 70-year-old man with claudication at half a mile and an 80-year-old man with claudication at half a mile or 100 m were significantly more likely to be referred (P < 0.001) in West Berkshire in 2000 compared with 1993. This referral change also applied to an 80-year-old man with an aortic aneurysm who was more likely to be referred in 2000 (P < 0.01). Patients with gradual onset of rest pain were more likely to be referred urgently or as an emergency in 2000 (P < 0.05). When comparing the two areas in 2000, significant differences emerged in the likelihood of referring patients with intermittent claudication. Scepticism continues towards the value of AAA screening in West Berkshire with only about two-thirds of GPs thinking that it was of value. On the other hand in Llantrisant, 94% of GPs thought AAA screening was valuable (P < 0.001). There was a significant decrease in the number of GPs who felt that they would refer patients directly to a regional vascular centre in West Berkshire between the two time periods (P < 0.001). There was a difference in likelihood of referral to regional centres between the two areas in 2000 (33% versus 6%, P < 0.001). CONCLUSIONS: Vascular referral patterns change with time and vary from one area to another. This has implications for planning vascular services.     

Direct optometrist referral of cataract patients into a pilot 'one-stop' cataract surgery facility

OBJECTIVES: To determine the feasibility of a) direct optometrist referral of patients with cataract, and b) combined assessment with same day cataract surgery ('one stop' cataract surgery). METHODS: Evaluation of 169 patients referred directly by optometrists into a pilot 'one stop' cataract surgery facility. RESULTS: Of 169 referrals, 160 patients (94.7%) were given confirmed appointments for the 'one stop' cataract service and 9 patients (5.3%) were appointed conventionally. Of 160 patients attending the 'one stop' cataract service, 154 patients (96.3%) underwent cataract surgery at the same visit, in 4 patients (2.5%) cataract surgery was indicated but deferred and in 2 patients (1.3%) cataract surgery was not indicated. The referral was supplemented with information regarding the patient's medical history forwarded by the general practitioner for 3 patients (1.8%). There were no systemic or sight-threatening complications. 151 patients (98.1%) achieved a visual acuity of 6/12 or better at a mean of 31 days post-operatively. CONCLUSION: Optometrists can accurately predict the need for cataract surgery and refer directly into a pilot 'one stop' cataract surgery facility, without the need for general practitioner involvement. 'One stop' cataract surgery is feasible; benefits to the patient include the abolition of the need to visit the general practitioner for consultation and referral, and the hospital for pre-assessment.     

Training general surgery residents to avoid postoperative cardiac events

BACKGROUND: Expertise in cardiac risk assessment takes years to acquire, but unnecessary cardiology consultation delays treatment and consumes scarce resources. METHODS: A retrospective review was performed of the cardiac work-up and postoperative events during 1 year on a general surgery service. Postgraduate year 1-3 general surgery residents were instructed to obtain a cardiology consult if a patient had any of the following: (1) had undergone coronary artery intervention more than 2 years in the past; (2) was taking an anti-anginal medication (nitroglycerine, Ca channel, or beta-blocker); or (3) was symptomatic or had an abnormal electrocardiogram. Whether a patient was symptomatic was to be tempered by the nature of the planned procedure. RESULTS: Supervised residents screened 720 unique patients for surgery. Cardiology consultation was obtained in 37. All but 1 (97%) patient referred to cardiology met at least 1 of the earlier-described criteria; with 8 (22%) meeting all 3 criteria. On average, patients referred to the cardiologists were taking 1.4 anti-anginal medications; and 1 patient sustained a fatal myocardial infarction after referral. Cardiac imaging (stress test or catheterization) was performed on 24 (65%) referred patients and was positive in 8 (33%). After minimizing cardiac risk by medication or intervention, the surgery service declined to offer the planned procedure to 11 (30%) of the referred patients and an additional 5 (15%) patients declined surgery. The overall surgical mortality was 2%. None of the patients in this series sustained a postoperative myocardial infarction or cardiac death. Postoperative supraventricular tachycardia was not influenced significantly by cardiology consultation (5% referred patients vs 1% nonreferred). CONCLUSIONS: Our criteria for obtaining cardiology consultation in general surgery patients appears to appropriately select patients in need of further work-up. Information obtained from a cardiac consultation frequently leads to a re-evaluation of the risks and benefits of surgery by both surgeons and patients.     

A comparison of stress and job satisfaction in female and male GPs and consultants

Questionnaires assessing sources and levels of job satisfaction and occupational stress were completed by male and female general practitioners (N = 547) and consultant doctors (N = 449) in Scotland during a period of structural change in the health service. Job satisfaction and occupational stress scales from the Occupational Stress Inventory³⁰ were used to compare GPs and consultants, males and females, and to enable comparisons with other occupational groups. Consultants had greater occupational stress than GPs on three OSI subscales detailing managerial aspects of their work, and also had greater job satisfaction overall than GPs. GPs were more stressed by ‘intrinsic’ factors of the job itself. Females had less occupational stress and greater job satisfaction overall than males. Male GPs as a group had greatest job stress and least job satisfaction. Younger principals in general practice and consultants reported more occupational stress and less job satisfaction than older doctors. These findings have implications for health service management in terms of providing support and training in stress management for younger doctors and for female doctors. Comparisons with combined sample norms for mainly professional and managerial occupations showed that GPs and consultants in the Scottish sample had generally greater job satisfaction and less occupational stress than norms, which is contrary to expectations.     

Effects of a shared protocol between urologists and general practitioners on referral patterns and initial diagnostic management of men with lower urinary tract symptoms in Italy: the Prostate Destination study

OBJECTIVE: To assess whether adopting a shared protocol between urologists and general practitioners (GPs) might change diagnostic procedures and referral patterns in the management of men with lower urinary tract symptoms (LUTS). SUBJECTS AND METHODS: Forty-five urological centres and 263 GPs in Italy participated in this prospective study. Procedures adopted by GPs for evaluating five consecutive patients (aged > or = 50 years) were compared before (phase 1) and after (phase 2) implementation of the shared protocol. An evidence-based diagnostic algorithm was developed and approved by participating urologists and presented to local GPs at a training session. Protocol modifications were allowed after discussion with GPs. Direct costs of diagnostic procedures carried out before and after implementing the protocol were calculated from the perspective of the national health service. RESULTS: In all, 903 patients were evaluable in phase 1 and 856 in phase 2. Implementation of the protocol did not change referral patterns, with about half the patients being managed entirely by GPs. The use of a digital rectal examination by GPs increased from 32% to 41%, use of transrectal and suprapubic ultrasonography decreased from 33% to 23% and 53% to 44%, respectively, (all P < 0.001) and use of the International Prostate Symptom Score increased from 4.5% to 23.1% (P < 0.001). Overall, protocol-recommended tests were used more frequently, while those not recommended decreased after implementing the protocol. However, overuse of the tests not recommended (i.e. urine culture and free/total prostate specific antigen ratio) remained high. The mean cost per patient of diagnostic procedures ordered by GPs decreased from Euros 71.82 to Euros 61.93, with Euros 9.9 saved for each patient. CONCLUSION: Our intervention failed to decrease the percentage of cases of LUTS being referred to specialists, but was moderately effective in inducing changes in the diagnostic management by GPs that were indicative of increased compliance with best-practice principles, and produced cost savings of 13.8%.     

First year of a fast track carotid duplex service.

BACKGROUND: The risk of major stroke is highest in the first three months after a transient ischaemic attack (TIA). Urgent carotid endarterectomy can reduce the risk in patients with a severe carotid stenosis. An express carotid duplex service has been established and this study analyses the effect in the first year. METHODS: Local GPs were offered a new service whereby a carotid duplex scan would be guaranteed within 14 days of referral of any patient who had a recent clearly documented TIA or amaurosis fugax. Referral letters were faxed directly to the Vascular Office where the duplex scan was authorised by a consultant vascular surgeon. Patients with significant carotid disease on duplex were assessed in the out-patient clinic in preparation for surgery. RESULTS: In the first 12 months of the service, 90 fast track duplex scans were performed. In the same interval 490 routine carotid duplex scans were carried out. Some 13 carotid endarterectomies were carried out on patients from the fast track group (13/90, 14%), with a median delay between referral and surgery of 30 days (range 20-45) and median duration between onset of symptoms and surgery of 51 days (range 27-406). In the non fast track group 14/490 (2.8%) scans resulted in carotid intervention, a median 127 days (range 64-184) after referral. CONCLUSION: The fast track service significantly reduced the delay between referral and surgery. Timing of carotid surgery is critical. Performing the surgery at the time of greatest risk increases the benefit of carotid endarterectomy. Urgent and appropriate referral from the GPs is vital for the service to run effectively.     

Utilization and outcome in the medical patient referred to surgery

The objective of this study was to test the hypothesis that hospitalized patients referred to a general surgical service from a medical service for a surgical procedure would have higher hospital costs and longer lengths of stay per diagnosis-related group (DRG) than patients admitted directly to the general surgical service. Hospital costs by DRG, exclusive of physician's fees, were analyzed for all adult general surgical admissions treated at our hospital from January 1, 1985 to March 31, 1986 (3,028 patients) to yield a population of patients in those DRGs with patients referred to general surgery from medicine (1,495 patients). Patients within each DRG were then disaggregated by either direct admission to general surgery (1,412 patients) or referral to the general surgical service from the medical service (83 patients). Mean cost per patient was 146.5 percent higher for referral patients than for direct admission patients, as was the total length of stay. Mortality was higher for referral patients than for direct admission patients. Factors analyzed which contributed to this greater resource utilization and higher mortality were (1) a greater severity of illness, (2) higher diagnostic costs, and (3) delays in diagnosis or treatment. The DRG payment for referral patients also produced a substantial deficit for the hospital, whereas direct admission patients produced a profit of +1,105,596. This data suggests that direct admission to the surgical service of patients likely to need surgery might lower their hospital costs and improve the quality of their care.     

Prostate-specific antigen testing in general practice: a survey among 325 general practitioners in Denmark

OBJECTIVES: Prostate-specific antigen (PSA) is a well-known and -utilized tumor marker for prostate cancer. Elevated PSA values are not specific for prostate cancer as they may be caused by other benign conditions. PSA testing is widely used by urologists and non-urologists. Interpretation of test results is difficult but important. Referral of patients for further work-up on suspicion of prostate diseases is mainly done by general practitioners (GPs). As the GP remains the gatekeeper between the patient and the urologist in terms of diagnosing prostate diseases, basic knowledge of PSA testing is crucial. The purpose of this study was to evaluate the basic use and knowledge of PSA testing and to give an estimate of the need for further education in PSA testing amongst GPs in our area. MATERIAL AND METHODS: A questionnaire regarding PSA testing and associated needs for education was mailed to all GPs in the Northern County of Denmark. Non-respondents were contacted by mail. RESULTS: Of the contacted GPs, 90% responded. Only 28% of GPs measured PSA in all males complaining of lower urinary tract symptoms (LUTS). Of patients seen as part of a general health check-up, PSA testing was done in 10%. The median PSA value for referral to urologists for further work-up was 5 ng/ml, but the decision was influenced by PSA value (79%), age (65%) and findings on digital rectal examination (DRE) (87%). Opportunistic screening for prostate cancer was done by 14% of GPs. Of the GPs who responded, 24% stated that they did not need any further education regarding PSA testing. CONCLUSIONS: The results of this study demonstrate that PSA testing is not standardized in our area. GPs do not test patients on the basis of recommendations provided by national or international societies. PSA testing is not used as a standard test in men with LUTS, and patients are not referred to urologists at a sufficiently low PSA level to improve the early diagnosis and work-up of patients with suspected prostate cancer. However, the decision of many GPs to refer patients to urologists for further work-up is influenced by the findings of a DRE, the age of the patient and the PSA value itself. Further education regarding PSA testing amongst GPs in our area would seem to be appropriate.     

Presumptive coding. A preliminary study of the aggregation of input diagnoses to determine their relevance to clinical governance and urological resourcing

OBJECTIVE: To determine the validity of a system for coding the reason for urological referral from primary care, using ICD10, and thus enable benchmarking of urological outpatient activity. PATIENTS AND METHODS: Four studies were conducted: (i) A pilot study to aggregate information into a few input diagnosis codes (925 patients); (ii) Validation of the aggregated codes using input diagnoses from a second centre (928 patients); (iii) A prospective study by three urologists to determine the system's generic utility (918 patients); (iv) A study to aggregate the presumptive codes for 2771 patients to gain an insight into the case-mix of patients referred to a general urological service via the outpatient department. RESULTS: The aggregation of input diagnoses from general practice referrals into 36 'presumptive codes' was possible and could be validated. Prospective coding, for 96% of eligible patients, was possible with < 1% of referral diagnoses not being codable. Further aggregation of the data for 2771 patients showed that 31% were referred with urological malignancy whilst 69% had symptoms suggesting benign urological disease. CONCLUSIONS: This preliminary study of presumptive coding suggests that it is a feasible and valid method of recording the input diagnoses for patients presented to a urological service. The information it provides has relevance for the structuring, benchmarking, resourcing and manpower requirements of that service, essential components for clinical governance. It also has relevance to the prospective collection of patient data for research and audit.     

Beurteilung des Verfahrens der Knieendoprothetik durch deutsche Hausärzte

To assess general knowledge and individual views of general practitioners (GPs) on total knee arthroplasty (TKA), we conducted a questionnaire-based survey involving 170 GPs in Germany. Eighty-one GPs returned the questionnaire. They treat a mean of 10.6+/-8.3 patients with TKA. General knowledge can be estimated as good. Compared with the data in the literature, GPs assessed the rate of satisfied patients as lower and the risk for revision surgery as higher. The mean risk of potential complications (infection, instability, persistent pain) in association with TKA was estimated correctly. Seventy-eight percent of GPs consider an allergic reaction to the implant or bone cement as problematic. The number of cases per year, personal experience with the surgeon, and - less important - local accessibility are important factors for GPs when recommending an operating centre to a patient.A broad spectrum of individual responses indicates the need to improve the information transfer between orthopaedic surgeons and referring GPs.     

Use of patient satisfaction data in a continuous quality improvement program for endoscopic sinus surgery.

OBJECTIVE: Continuous quality improvement (CQI) is an effort by health care providers to improve the quality of service by continuously exceeding patients' expectations. Patient satisfaction is one of the measures of the quality of care. The aims of this study were to report the patients' evaluation of endoscopic sinus surgery (ESS) and to explore the feasibility in using patient satisfaction data in the CQI program for ESS. METHODS: Eighty-three patients completed a validated patient satisfaction survey (PSS) 1 month after undergoing ESS. Logistic regression models were applied to determine the confounders of patient satisfaction. RESULTS: In general, 72% of patients were very satisfied with the services. Education level and milder disease correlated with higher overall satisfaction levels (P 相似文献   

The One-stop trial: Does electronic referral and booking by the general practitioner (GPs) to outpatient day case surgery reduce waiting time and costs? A randomized controlled trial protocol

Background

Waiting time and costs from referral to day case outpatient surgery are at an unacceptably high level. The waiting time in Norway averages 240 days for common surgical conditions. Furthermore, in North Norway the population is scattered throughout a large geographic area, making the cost of travel to a specialist examination before surgery considerable. Electronic standardised referrals and booking of day case outpatient surgery by GPs are possible through the National Health Network, which links all health care providers in an electronic network. New ways of using this network might reduce the waiting time and cost of outpatient day case surgery.

Materials and Methods

In a randomised controlled trial, selected patients (inguinal hernia, gallstone disease and pilonidal sinus) referred to the university hospital are either randomised to direct electronic referral and booking for outpatient surgery (one stop), or to the traditional patient pathway where all patients are seen at the outpatient clinic several weeks ahead of surgery. Consultants in gastrointestinal surgery designed standardised referral forms and guidelines. New software has been designed making it possible to implement referral forms, guidelines and patient information in the GP's electronic health record. For "one-stop" referral, GPs must provide mandatory information about the specific condition. Referrals were linked to a booking system, enabling the GPs to book the hospital, day and time for outpatient surgery. The primary endpoints are waiting time and costs. The sample size calculation was based on waiting time. A reduction in waiting time of 60 days (effect size), 25%, is significant, resulting in a sample size of 120 patients in total.

Discussion

Poor communication between primary and secondary care often results in inefficiencies and unsatisfactory outcomes. We hypothesised that standardised referrals would improve the quality of information, making it feasible to use a one-stop approach for all patients undergoing surgery on an outpatient basis for inguinal hernia, pilonidal sinus and gallstones. In this study we wanted to investigate the waiting time and cost-effectiveness of direct electronic referral and booking of outpatient surgery compared to the traditional patient pathway, where the patient is seen at the outpatient clinic prior to surgery.

Trial registration

This trial has been registered at ClinicalTrials.gov. The trial registration number is: NCT00692497