작성자가 등록한 다른 큐
댓글을 작성하기 위해서는 먼저
로그인
하셔야 합니다.
현재 댓글의 수는
0
개 입니다.
Page 0:
Page 1: Substitution of doctors by nurses in primary care (Review)
Laurant M, Reeves D, Hermens R, Braspenning J, Grol R, Sibbald B
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2007, Issue 3 http://www.thecochranelibrary.com
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
1
Page 2: TABLE OF CONTENTS ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW . . . . . . . . . . . . . . . . . . SEARCH METHODS FOR IDENTIFICATION OF STUDIES . . . . . . . . . . . . . . . . . . . METHODS OF THE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DESCRIPTION OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODOLOGICAL QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . POTENTIAL CONFLICT OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Characteristics of included studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Characteristics of excluded studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 01. Methodological Quality of controlled before and after studies . . . . . . . . . . . . . . . . Table 02. Methodological quality of randomized controlled trials . . . . . . . . . . . . . . . . . . Table 03. Patient outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 04. Process of Care Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 05. Resource Utilisation Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . Table 06. Cost Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison 01. Doctor-Nurse substitution study results . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COVER SHEET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GRAPHS AND OTHER TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 01. Trial Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 01.01. Comparison 01 Doctor-Nurse substitution study results, Outcome 01 Physical function (better vs not better) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 01.02. Comparison 01 Doctor-Nurse substitution study results, Outcome 02 Patient satisfaction . . . . Analysis 01.03. Comparison 01 Doctor-Nurse substitution study results, Outcome 03 Scheduled return visits . . . Analysis 01.04. Comparison 01 Doctor-Nurse substitution study results, Outcome 04 Prescription ordered . . . . Analysis 01.05. Comparison 01 Doctor-Nurse substitution study results, Outcome 05 Hospital referral . . . . . Analysis 01.06. Comparison 01 Doctor-Nurse substitution study results, Outcome 06 Attendance at Accident&Emergency Analysis 01.07. Comparison 01 Doctor-Nurse substitution study results, Outcome 07 Hospital admission . . . . 1 2 2 3 3 4 5 6 6 6 8 9 9 9 9 10 13 13 19 20 20 20 21 24 26 27 28 28 29 29 30 30 31 31 32 32 33 33 34
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
i
Page 3: Substitution of doctors by nurses in primary care (Review)
Laurant M, Reeves D, Hermens R, Braspenning J, Grol R, Sibbald B
This record should be cited as: Laurant M, Reeves D, Hermens R, Braspenning J, Grol R, Sibbald B. Substitution of doctors by nurses in primary care. Cochrane Database of Systematic Reviews 2004, Issue 4. Art. No.: CD001271. DOI: 10.1002/14651858.CD001271.pub2. This version first published online: 18 October 2004 in Issue 4, 2004. Date of most recent substantive amendment: 16 July 2004
ABSTRACT Background Demand for primary care services has increased in developed countries due to population ageing, rising patient expectations, and reforms that shift care from hospitals to the community. At the same time, the supply of physicians is constrained and there is increasing pressure to contain costs. Shifting care from physicians to nurses is one possible response to these challenges. The expectation is that nurse-doctor substitution will reduce cost and physician workload while maintaining quality of care. Objectives Our aim was to evaluate the impact of doctor-nurse substitution in primary care on patient outcomes, process of care, and resource utilisation including cost. Patient outcomes included: morbidity; mortality; satisfaction; compliance; and preference. Process of care outcomes included: practitioner adherence to clinical guidelines; standards or quality of care; and practitioner health care activity (e.g. provision of advice). Resource utilisation was assessed by: frequency and length of consultations; return visits; prescriptions; tests and investigations; referral to other services; and direct or indirect costs. Search strategy The following databases were searched for the period 1966 to 2002: Medline; Cinahl; Bids, Embase; Social Science Citation Index; British Nursing Index; HMIC; EPOC Register; and Cochrane Controlled Trial Register. Search terms specified the setting (primary care), professional (nurse), study design (randomised controlled trial, controlled before-and-after-study, interrupted time series), and subject (e.g. skill mix). Selection criteria Studies were included if nurses were compared to doctors providing a similar primary health care service (excluding accident and emergency services). Primary care doctors included: general practitioners, family physicians, paediatricians, general internists or geriatricians. Primary care nurses included: practice nurses, nurse practitioners, clinical nurse specialists, or advanced practice nurses. Data collection and analysis Study selection and data extraction was conducted independently by two reviewers with differences resolved through discussion. Metaanalysis was applied to outcomes for which there was adequate reporting of intervention effects from at least three randomised controlled trials. Semi-quantitative methods were used to synthesize other outcomes. Main results 4253 articles were screened of which 25 articles, relating to 16 studies, met our inclusion criteria. In seven studies the nurse assumed responsibility for first contact and ongoing care for all presenting patients. The outcomes investigated varied across studies so limiting the opportunity for data synthesis. In general, no appreciable differences were found between doctors and nurses in health outcomes for patients, process of care, resource utilisation or cost. In five studies the nurse assumed responsibility for first contact care for patients wanting urgent consultations during office hours or out-of-hours. Patient health outcomes were similar for nurses and doctors but patient satisfaction was higher with nurse-led care. Nurses tended to provide longer consultations, give more information to patients and recall patients more frequently than did doctors. The impact on physician workload and direct cost of care was variable.
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd 1
Page 4: In four studies the nurse took responsibility for the ongoing management of patients with particular chronic conditions. The outcomes investigated varied across studies so limiting the opportunity for data synthesis. In general, no appreciable differences were found between doctors and nurses in health outcomes for patients, process of care, resource utilisation or cost. Authors’ conclusions The findings suggest that appropriately trained nurses can produce as high quality care as primary care doctors and achieve as good health outcomes for patients. However, this conclusion should be viewed with caution given that only one study was powered to assess equivalence of care, many studies had methodological limitations, and patient follow-up was generally 12 months or less. While doctor-nurse substitution has the potential to reduce doctors’ workload and direct healthcare costs, achieving such reductions depends on the particular context of care. Doctors’ workload may remain unchanged either because nurses are deployed to meet previously unmet patient need or because nurses generate demand for care where previously there was none. Savings in cost depend on the magnitude of the salary differential between doctors and nurses, and may be offset by the lower productivity of nurses compared to doctors.
PLAIN LANGUAGE SUMMARY In primary care, it appears that appropriately trained nurses can produce as high quality care and achieve as good health outcomes for patients as doctors. However, the research available is quite limited. Many countries have sought to shift the provision of primary care from doctors to nurses in order to reduce the demand for doctors and improve healthcare efficiency. The expectation is that nurses working as substitutes can provide as high quality care as doctors at lower cost. This review found that quality of care is similar for nurses and doctors but it is not known if it decreases the doctor’s workload. Nurses tend to provide more health advice and achieve higher levels of patient satisfaction compared with doctors. Even though using nurses may save salary costs, nurses may order more tests and use other services which may decrease the cost savings of using nurses instead of doctors.
BACKGROUND Demand for primary care services has increased in many countries due to population ageing, rising patient expectations, and reforms that shift care from hospitals to the community. At the same time, the supply of physicians is constrained and there is increasing pressure to contain costs. Shifting care from doctors to nurses, is one possible response to these challenges (Jenkins-Clarke 1998; Whitecross 1999). A review of research into the substitutability of nurses for doctors suggested that 25% to 70% of the work undertaken by doctors might be moved to nurses (Richardson 1998). In primary care, nurses may undertake much of the health promotion work of family practice (Family HSG 1994; Muir 1995), and play a leading role in the routine management of chronic diseases such as asthma, diabetes and coronary heart disease (Aubert 1998; Charlton 1991; Kirkman 1994). The expectation is that primary care nurses working in extended roles can: a)enhance the quality of services provided by doctors; b)safely substitute for doctors in an wide array of services, so reducing demand for doctors; and c)reduce the direct costs of services because nurse are cheaper to hire than physicians. Nurses may work either as doctor supplements or as doctor sub-
stitutes. Nurses working as doctor supplements provide services which complement or extend those provided by doctors. The aim is to improve the quality of care and extend the range of services available to patients. In contrast, nurses working as doctor substitutes provide services which otherwise would be provided by doctors alone. The aim is to reduce the demand for doctors. Gains in service efficiency may be achieved if doctors give up providing the services they have delegated to nurses, and instead invest their time in activities that only doctors can perform (Richardson 1999). This review is focused on the impact of nurses working as substitutes for primary care doctors. Previous systematic reviews of doctor-nurse substitution in primary care have sought to identify whether nurses differ from doctors in terms of patient outcomes, process of care or resource utilisation. In 1995, Brown and Grimes conducted a meta-analysis of American and Canadian research into doctor-nurse substitution in primary care (Brown 1995). Thirty-eight studies were included in the review, covering a wide range of nursing roles and encompassing both (quasi) experimental and observational research designs. The findings suggested that, as compared with doctor-led care, nurse-led care was associated with higher levels of patient compliance and satisfaction, longer consultations, and higher rates of laboratory testing. Health outcomes for patients were similar.
2
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 5: These findings are supported by the more recent systematic review of Horrocks et al in 2002 (Horrocks 2002). They included 11 randomised controlled trials and 23 prospective observational studies of nurses acting as doctor substitutes for patients with undifferentiated healthcare problems in primary care settings in developed countries. The findings suggested that patient health care outcomes were similar for doctors and nurses, but that nurse-led care was associated with higher levels of patient satisfaction, longer consultations and higher rates of investigation. Both reviews are flawed in their inclusion of observational research which is susceptible to producing biased estimates of differences between doctors and nurses through failure to control for other factors that may affect outcome. In addition, by combining a diversity of nurse roles, it remains unclear whether the observed differences or similarities between nurses and doctors vary with the particular type of role substitution. We aimed to address these deficiencies.
• Nurses - any qualified nurse working as a substitute to a primary care physician. This could include: nurse practitioners, clinical nurse specialists, advanced practice nurses, practice nurses, health visitors, etc. As the job title, education, and experience of nurses varies considerably among and within countries, we did not select nurses by virtue of their job title. Only trainee nurses and mental health nurses were excluded. • Patients - presenting in primary care, excluding accident and emergency. The review is limited to primary health care services that provide first contact and ongoing care for patients with all types of health problems. It includes family practice or general practice, outpatient settings, and ambulatory primary care settings (excluding accident and emergency). Types of intervention Our focus was on nurses working as substitutes for primary care doctors. Substitution refers to the situation where task(s) formerly performed by one type of professional (i.e. doctor) are transferred to a different type of professional (i.e. nurse), usually with the intention of reducing cost or addressing workforce shortages. Substitution studies typically examine the case where a nurse is responsible for providing the same health care as a doctor, and the performance of these two practitioners is compared. For example, a nurse-led clinic for a particular disease or condition is compared to a doctor-led clinic. Supplementation refers to the situation where a nurse supplements or extends the care of the doctor by providing a new primary care service. The aim is generally to improve the quality of care rather than reduce cost or address workforce shortages. Supplementation studies typically compare usual care by a doctor to an innovative service provided by a nurse working alongside a doctor. For example, a family practice with a nurse-led diabetes clinic is compared to a family practice without such a clinic. This type of study risks confounding two aspects of care provision: a)type of service (specialised clinic vs routine consultation), and b)who provides that service (doctor or nurse) Supplementation studies have been excluded from this review. Some studies investigated complex interventions where practitioner care was combined with other interventions. Where a factorial study design was employed, the nurse can be compared with the doctor independently of other interventions. In such cases, we report only the effect attributable to the nurse compared with the doctor. Types of outcome measures Four types of outcomes were considered for this review, patient outcomes, process of care outcomes, resource utilisation outcomes and cost outcomes. Patient outcomes • Morbidity.
3
OBJECTIVES Our aim was to investigate the impact of nurses working as substitutes for primary care doctors on: • Outcomes for patients • Process of care • Resource utilisation • Direct (service) and indirect (societal) costs
CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW Types of studies Three types of study were eligible for inclusion: • Randomised controlled trials (RCT): Random or quasi-random allocation of subjects to intervention and control groups. • Controlled before and after studies (CBA): the intervention group is compared with a control group selected by non-random processes. Outcomes must be measured before as well as after the intervention. • Interrupted time series (ITS). Longitudinal examination of outcomes with at least three observations before and again after the intervention. Types of participants • Doctors - primary care physicians which could include general practitioners, family doctors, paediatricians, general internists or geriatricians.
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 6: • Mortality. • Quality of life. • Satisfaction. • Patient compliance. • Other (knowledge, preference for doctor or nurse). Process of care outcomes • Practitioner adherence to clinical guidelines. • Standards or quality of care. • Practitioner health care activity (examinations, provision of advice). Resource utilisation outcomes • Frequency and length of consultations. • Return visits. • Prescriptions. • Tests and investigations. • Referral/use of other services. Cost outcomes • Direct (service) . • Indirect (societal) costs.
SEARCH METHODS FOR IDENTIFICATION OF STUDIES See: Cochrane Effective Practice and Organisation of Care Group methods used in reviews. The following databases were searched: Medline; Cinahl; Bids Embase; Social Science and Citation Indexes; British Nursing index; HMIC; EPOC Register; Cochrane Controlled Trial Register (CCTR); and the National Primary Care Research and Development Centre’s own database. The search terms combined Medical Subject Headings (MeSH) and free text words as shown in the strategy below. The search was conducted first in 1999 (1966 till 1999) and then updated in 2002 (1999 till 2002). The updated search was restricted by study design. Search strategy #1 (’Family-Practice’ / all topical subheadings / all age subheadings in DE) or (’Physicians-Family’ / all topical subheadings / all age subheadings in DE) or (’Primary-HealthCare’ / all topical subheadings / all age subheadings in DE) or (’Primary-Nursing’ / all topical subheadings / all age subheadings in DE) or (’Nursing-Care’ / all topical subheadings / all age subheadings in DE) #2 primary near care
#3 #1 or #2 #4 (’Community-Health-Nursing’ / all topical subheadings / all age subheadings in DE) or (’Nurse-Administrators’ / all topical subheadings / all age subheadings in DE) or (’Nurse-Midwives’ / all topical subheadings / all age subheadings in DE) or (’NursePractitioners’ / all topical subheadings / all age subheadings in DE) or (’Clinical-Nurse-Specialists’ / all topical subheadings / all age subheadings in DE) #5 nurse* #6 nurse manager* #7 district nurs* #8 practice nurs* #9 health visit* #10 #4 or #5 or #6 or #7 or #8 or #9 #11 (’Cooperative-Behavior’ / all topical subheadings / all age subheadings in DE) or (’Job-Description’ / all topical subheadings / all age subheadings in DE) or ( ’ProfessionalAutonomy’ / all topical subheadings / all age subheadings in DE) or (’Clinical-Competence’ / all topical subheadings / all age subheadings in DE) #12 clinical practice #13 deleg* #14 multidisplin* #15 substitut* #16 cooperat* #17 role* #18 skill mix #19 health promotion #20 team* #21 patient counselling #22 nurs* near5 general pract* #23 #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 #24 (’Clinical-Trials’ / all topical subheadings / all age subheadings in DE) or (’Double-Blind-Studies’ / all topical subheadings / all age subheadings in DE) or (’EvaluationResearch’ / all topical subheadings / all age subheadings in DE) or (’Prospective-Studies’ / all topical subheadings / all age subheadings in DE) or (’Single-Blind-Studies’ / all topical subheadings / all age subheadings in DE) or (’Study-Design’ / all topical subheadings / all age subheadings in DE) #25 random allocation #26 rct #27 (randomised controlled trial*) or (randomized controlled trial*) #28 comparative stud* #29 interupted time series #30 #24 or #25 or #26 or #27 or #28 or #29 #31 #3 and #10 and #23 and #30 with limitations #32 #31 and (PY=1999-2001) and (DT=JOURNAL-ARTICLE) and (LA=ENGLISH)
4
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 7: Medline SEARCH STRATEGY UPDATE: Combination Set 1 + Set 2 + Set 3 + Set 4 (with limitations) #31 #30 and (PY=1999-2001) and (LA=ENGLISH) and (PT= JOURNAL-ARTICLE) (59 records) #30 #3 and #10 and #23 and #29 (95 records) Set 4: Design/methodology #29 #24 or #25 or #26 or #27 or #28 (141427 records) #28 interupted time series (0 records) #27 Comparative stud* (4656 records) #26 (randomised controlled trial*) or (randomized controlled trial*) (14922 records) #25 rct (350 records) #24 (’Clinical-Trials’ / all subheadings in MIME,MJME) or (’Controlled-Clinical-Trials’ / all subheadings in MIME,MJME) or (’Double-Blind-Method’ / all subheadings in MIME,MJME) or (’Evaluation-Studies’ / all subheadings in MIME,MJME) or (’Follow-Up-Studies’ / all subheadings in MIME,MJME) or (’Prospective-Studies’ / all subheadings in MIME,MJME) or (’Random-Allocation’ / all subheadings in MIME,MJME) or (’Randomized-Controlled-Trials’ / all subheadings in MIME,MJME) or (’Single-Blind-Method’ / all subheadings in MIME,MJME) or (’Research-Design’ / all subheadings in MIME,MJME) (134715 records) Set 3: Substitution #23 #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 (255757 records) #22 nurs* near5 general pract* (156 records) #21 patient counseling (263 records) #20 team* (11973 records) #19 health promotion (6052 records) #18 skill mix (53 records) #17 role* (189215 records) #16 cooperat* (14653 records) #15 substitut* (30365 records) #14 multidisplin* (4 records) #13 deleg* (428 records) #12 clinical practice (8293 records) #11 (’Cooperative-Behavior’ / all subheadings in MIME,MJME) or (’Job-Description’ / all subheadings in MIME,MJME) or (’Professional-Autonomy’ / all subheadings in MIME,MJME) or (’Clinical-Competence’ / all subheadings in MIME,MJME) (10749 records) Set 2: Nurse #10 #4 or #5 or #6 or #7 or #8 or #9 (50752 records) #9 health visit* (335 records) #8 practice nurs* (2311 records) #7 district nurs* (110 records) #6 nurse manager* (192 records) #5 nurs* (50649 records) #4 (’Nurse-Administrators’ / all subheadings in MIME,MJME) or (’Nurse-Clinicians’ / all subheadings in MIME,MJME)
or (’Nurse-Midwives’ / all subheadings in MIME,MJME) or (’Nurse-Practitioners’ / all subheadings in MIME,MJME) or (’Community-Health-Nursing’ / all subheadings in MIME,MJME) (6662 records) Set 1: Setting #3 #1 or #2 (18096 records) #2 primary near care (12217 records) #1 (’Family-Practice’ / all subheadings in MIME,MJME) or (’Primary-Health-Care’ / all subheadings in MIME,MJME) or (’Primary-Nursing-Care’ / all subheadings in MIME,MJME) or (’Physicians-Family’ / all subheadings in MIME,MJME) (12588 records) The titles and abstracts of articles uncovered by the above searches were independently screened by two reviewers. The full text of potentially relevant articles was obtained for further evaluation. The reference lists of included articles, and of existing published reviews of doctor-nurse substitution, were checked for other potentially relevant studies. Only articles written in English or Dutch were included.
METHODS OF THE REVIEW Each potentially relevant study was independently assessed for inclusion in the review by two reviewers. Differences between the reviewers were resolved by discussion. Excluded studies along with the reasons for their exclusion are given below (See Reference list, excluded studies). A data extraction form based on the standard EPOC checklist was designed for this review (See Group Details). Data from each included study were extracted independently by two reviewers. Differences were resolved by discussion. If a single publication reported two or more separate studies, then each study was extracted separately. If the findings of a single study were spread across two or more publications, then the publications were extracted as one. For each study with more than one control or comparison group for the nurse intervention, we report only the results for the control condition in which doctors provided the same intervention as the nurse. Standard EPOC criteria were used to assess the methodological quality of the studies (See Assessment of Methodological Quality in Group Details). Analysis Studies were grouped by nurse role for analysis, as follows: • First contact and ongoing care for undifferentiated patients. • First contact care for patients wanting urgent attention during office hours or out-of-hours. • Routine management of patients with chronic conditions. For each group, meta-analysis was applied to outcomes for which there was adequate reporting of intervention effects from
5
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 8: at least three randomised controlled trials. We excluded nonrandomised studies from meta-analysis due to their inherently greater potential for bias, and we excluded outcomes for which less than three randomised controlled trials were available on the grounds that a meta-analysis would not add substantial value to a semi-quantitative examination. Outcomes not amenable to metaanalysis were subjected to semi-quantitative synthesis. All results reported below are statistically significant unless otherwise stated. A fixed-effects (FE) model was used for all meta-analyses. A general recommendation is that, when there is evidence of substantial heterogeneity between study results, a Random Effects (RE) model should also be applied (Petitti 2001). Despite significant heterogeneity, we decided against the use of RE models for two main reasons: (i) we had very small numbers of studies in each analysis - three at most; and (ii) we had no basis for assuming that effect sizes are normally distributed. These factors can result in RE estimates of overall effect and confidence intervals that are quite inaccurate (Alderson 2004; Biggerstaff 1997; Maas 2004) and, in our view, the risk of drawing an inappropriate conclusion from the analysis was too large to justify the use of RE models. In contrast, the FE model is known to provide accurate estimates of the average effect (and confidence interval) within the included studies even when the number of studies is small(Brockwell 2001), but does not provide a statistical basis for generalising beyond the studies in hand (Bailey 1987).
METHODOLOGICAL QUALITY Of the 16 studies included, three were controlled before-and-after studies (Chambers 1977, Gordon 1974, Myers 1997) and 13 were randomised or quasi-randomised controlled trials (Chambers 1978; Flynn 1974; Hemani 1999; Kinnersley 2000; Lattimer 1998; Lewis 1969; McIntosh 1997; Moher 2001; Mundinger 2000; Shum 2000; Spitzer 1973; Stein 1974; Venning 2000) (See Characteristics of Included Studies). The methodological quality of controlled before-and-after studies (Chambers 1977; Gordon 1974; Myers 1997) was assessed by nine quality criteria (Table 01). None of the three controlled before-and-after studies reported the statistical power. In one study (Chambers 1977) the unit of allocation was the community, whereas the unit of analysis was the patient with no allowance for clustering. In all three studies the intervention and control groups appeared to be comparable. Each study fulfilled four of the nine quality criteria. The methodological quality of randomised controlled trials (Chambers 1978; Flynn 1974; Gordon 1974; Hemani 1999, Kinnersley 2000, Lattimer 1998, Lewis 1969, Moher 2001, Mundinger 2000, Shum 2000, Spitzer 1973, Stein 1974, Venning 2000) was also assessed by nine criteria (Table 02). All studies had methodological shortcomings. The power was reported in five of 13 trials (Kinnersley 2000; Lattimer 1998; McIntosh 1997; Mundinger 2000; Shum 2000), and two studies (Chambers 1978; Hemani 1999) reported that the study lacked the statistical power to detect clinically meaningful differences. Two studies (Chambers 1978; Spitzer 1973) used cluster randomisation without correcting for clustering in the analysis. Concealment of allocation was not reported in seven studies (Chambers 1978; Flynn 1974; Lewis 1969; McIntosh 1997; Moher 2001; Mundinger 2000; Spitzer 1973). In 12 out of 13 trials (Chambers 1978; Flynn 1974; Gordon 1974; Hemani 1999; Kinnersley 2000; Lattimer 1998; Lewis 1969; McIntosh 1997; Mundinger 2000; Myers 1997; Shum 2000; Spitzer 1973; Stein 1974; Venning 2000) it was unclear whether or not contamination had occurred. Of the 13 trials, none fulfilled eight or more criteria; seven studies met four to seven criteria; and six studies met three or fewer criteria.
DESCRIPTION OF STUDIES The initial searches identified 4253 potentially relevant articles (3784 in the original search, and 469 in the updated search) of which 25 publications, relating to 16 studies met our inclusion criteria (Figure 01). We identified a further 46 studies which investigated the role of nurses working as supplements to primary care doctors; these have been excluded from the review and will be reported separately. We included 16 studies. In seven studies ( Chambers 1977; Chambers 1977; Flynn 1974; Hemani 1999, Mundinger 2000,Spitzer 1973), the nurse assumed responsibility for first contact and ongoing care for all presenting patients. In five studies the nurse assumed responsibility for first contact care for patients wanting urgent consultations during routine practice hours (Kinnersley 2000; Myers 1997; Shum 2000; Venning 2000) or out-of-hours (Lattimer 1998). In four studies (Lewis 1969; McIntosh 1997; Moher 2001; Spitzer 1973) the nurse had responsibility for the ongoing management of patients with particular chronic conditions. In one of these studies (McIntosh 1997) the nurse provided counselling to problem drinkers. In all studies, the control or comparison group consisted of doctors providing the same services to patients as the nurses.
RESULTS A. First contact and ongoing care for all presenting patients. Patient outcomes were assessed in five studies (Chambers 1978; Flynn 1974; Gordon 1974; Mundinger 2000; Spitzer 1973). Health status was investigated in four of these (Chambers 1978; Gordon 1974; Mundinger 2000; Spitzer 1973); 25 outcomes were measured of which two were significantly better with nurse-led care and 23 showed no significant difference. One of the two observed differences between nurses and doctors is untrustworthy in that the study (Chambers 1978) made no allowance for cluster randomisation in the analysis. Patient satisfaction was assessed
6
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 9: in three studies (Gordon 1974; Mundinger 2000; Spitzer 1973); 15 outcomes were measured of which one was significantly better with doctor-led care and 14 showed no significant difference. Patient compliance was assessed in two studies (Flynn 1974; Gordon 1974); four outcomes were measured and none differed significantly between doctors and nurses. Patient knowledge was assessed in one study; three outcomes were measured of which one was significantly better with nurse-led care and two showed no significant difference (Table 03). Process of care was assessed in four studies (Chambers 1977; Flynn 1974; Gordon 1974; Spitzer 1973)). Of the 12 outcomes measured, three were significantly better with nurse-led care. In 2 cases the nurse was significantly more likely than the doctor to provide lifestyle advice (Flynn 1974). In one case, sub-group analysis suggested that the nurse had significantly fewer lapses in care when treating patients with unstable chronic disease (Gordon 1974). The remaining nine outcomes showed no significant difference (Table 04). Resource utilisation was assessed in three studies (Flynn 1974; Hemani 1999; Mundinger 2000). Consultation rates were investigated in two studies (Hemani 1999; Mundinger 2000) and neither found a significant difference between doctors and nurses. Tests and investigations were assessed in two studies (Flynn 1974; Hemani 1999); 22 outcomes were measured of which four showed significantly higher rates for nurses and the remainder showed no difference. Use of other health care services was assessed in all three studies; seven outcomes were measured of which one showed a significantly higher rate for nurses and the remainder showed no significant difference (Table 05). Direct costs were assessed in two studies (Chambers 1977; Spitzer 1973) and no significant differences were found (Table 06). B. First contact care for patients wanting urgent attention. Patient outcomes were assessed in four studies (Kinnersley 2000; Lattimer 1998; Shum 2000; Venning 2000). Health status was investigated in all four studies; five outcomes were measured and none differed significantly between doctors and nurses. Patient satisfaction was assessed in 3 studies (Kinnersley 2000; Shum 2000; Stein 1974); 19 outcomes were measured of which 12 were significantly better with nurse-led care and seven showed no significant difference. Meta-analysis of three studies (Kinnersley 2000; Shum 2000; Venning 2000) showed that patient satisfaction was higher with nurse-led care as compared with doctor-led care (standardised mean difference 0.28, 95% confidence interval:0.21, 0.34) but the effect size was highly variable between studies (See Comparison 01.02). Patient compliance and enablement were measured in one study (Venning 2000) and no significant differences were found (Table 03). Process of care was assessed in three studies (Kinnersley 2000; Shum 2000; Venning 2000). Of the eight outcomes measured, six were significantly better with nurse-led care - all showing nurses
provided more information to patients than did doctors. The remaining two outcomes showed no significant difference (Table 04). Resource utilisation was assessed in five studies (Kinnersley 2000; Lattimer 1998; Myers 1997; Shum 2000; Venning 2000). Consultation length was measured in three studies (Kinnersley 2000; Shum 2000; Venning 2000) and all showed significantly longer consultations for nurses. Consultation rate was investigated in three studies (Kinnersley 2000; Shum 2000; Venning 2000); five outcomes were measured of which three showed significantly higher rates for nurses and the remainder showed no difference. Meta-analysis of these three studies showed that nurses were more likely than doctors to recall a patient (relative risk nurse will recall patient compared with doctor = 1.34, 95% confidence interval: 1.20, 1.49) but with considerable heterogeneity across studies (See Comparison 01.03). Prescribing rate was investigated in three studies (Kinnersley 2000; Shum 2000; Venning 2000); five outcomes were measured of which one showed a lower rate for nurses and the remainder showed no difference. Meta-analysis of these three studies suggested there was no significant difference between doctors and nurses in prescribing rates (relative risk nurse will prescribe compared to doctor = 1.00, 95% confidence limit: 0.96, 1.05) (See Comparison 01.04). Tests and investigations were examined in two studies (Kinnersley 2000; Venning 2000); two outcomes were measured of which one showed a higher rate for nurses. Use of other services was investigated in five studies (Kinnersley 2000; Lattimer 1998; Myers 1997; Shum 2000; Venning 2000); nine outcomes were measured and none showed a significant difference between nurses and doctors. Meta-analysis of three studies (Kinnersley 2000; Lattimer 1998; Venning 2000) suggested that there were no significant differences between doctors and nurses in referral rates to hospital (relative risk of referral by nurse compared to doctor = 0.79, 95% confidence interval: 0.58, 1.07) (Comparison 01.05). Doctors’ workload was assessed in one study (Lattimer 1998); three outcomes were assessed, all of which showed a reduction in doctors’ workload with nurse-led care (Table 05). Costs were assessed in two studies (Lattimer 1998; Venning 2000). One study (Lattimer 1998) showed a net reduction in direct costs with nurse-led care while the other (Venning 2000) found no difference (Table 06). C. Routine management of patients with chronic conditions. Patient outcomes were assessed in four studies (Lewis 1969; McIntosh 1997; Moher 2001; Stein 1974). Health status was assessed in all four studies; eight outcomes were measured of which one was significantly better with nurse-led care and seven showed no significant difference. Patient satisfaction was assessed in one study (Lewis 1969) and was found to be significantly higher with nurseled care. Compliance was assessed in one study (Lewis 1969) and no significant difference was found. Patient knowledge was as7
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 10: sessed in one study (Stein 1974) and was found to be significantly higher with nurse-led care (Table 03). Process of care was investigated in one study (Moher 2001). Of the four outcomes measured, none differed significantly (Table 04). Resource utilisation was assessed in two studies (Moher 2001; Stein 1974). Consultation rate was examined in one study (Stein 1974) and no significant difference was found. Prescribing rates were investigated in both studies; four outcomes were measured and none showed a significant difference (Table 05). Direct cost of care was assessed in one study (Lewis 1969) and no significant difference was found (Table 06).
(Shum 2000) did not. Moreover, no appreciable differences were found between doctors and nurses in other aspects of resource utilisation such as prescribing, use of tests or investigations, or referrals to other services. It therefore seems unlikely that the lower productivity of nurses as compared with doctors reflects their relative inexperience. While no appreciable differences in resource use were found when nurses substituted for doctors in providing ongoing care for undifferentiated patients or those with particular chronic conditions, caseload (number of patients seen per unit of time) was not measured so productivity is unknown. Only one of five studies (Lattimer 1998) in which nurses provided first contact care for patients wanting urgent attention out-ofhours, demonstrated clear cost savings with nurse-led services. In all other studies - spanning all three of the nursing roles considered in this review - the lower salary costs of nurses were offset by their increased use of resources or lower productivity. As salary differentials between nurses and doctors may vary from place to place and over time, the net saving to health care services, if any, will be highly context dependent (Richardson 1998). Only 1 study investigated the impact of nurses on doctors’ workload and this showed reductions in the demand for doctors (Lattimer 1998) in which nurses provided first contact care for patients wanting urgent attention out-of-hours). However, a recent controlled trial of adding nurses to doctors’ teams showed no reduction in physician workload(Laurant 2004). This may be because nurses addressed previously unmet need or because nurses generated demand where previously there was none. In either case, the findings suggest that the addition of nurses to physician teams may not reduce workload unless active steps are taken to ensure doctors discontinue providing the services that have been transferred to nurses. Efficiency gains are possible if doctors invest this ’saved’ time in activities that only doctors can perform (Richardson 1999). It is self-evident that nurses must be adequately trained to act as substitutes for doctors. There is, however, no agreement as to the level of training required for nurses to undertake the specific roles covered by this review, and no consistency in the qualifications nurses must have to merit job titles such as nurse practitioner. Few studies contained detailed information on the nature of nurses’ training for the specific role under investigation, making it impossible for us to draw any conclusions as to whether or how training affects outcomes. All the studies included in this review adopted the position that the nurses they investigated were competent to carry out the clinical role assigned to them and, indeed, the evidence supports that assumption. Additional research is therefore needed to examine the relationship between training and outcome. The limitations of this review need to be considered. Our search strategy was designed to maximise sensitivity (detection of relevant research) at the expense of specificity (excluding irrelevant
8
DISCUSSION The findings suggest that nurses and doctors generate similar health outcomes for patients, at least in the short-term, over the range of care investigated. This work included the provision of first contact and/or ongoing care for unselected patients and the management of patients with specific chronic conditions. The findings must be viewed with caution, however, given that only 1 study (in which nurses provided first contact care for patients wanting urgent attention out-of-hours) was powered to assess equivalence of care. Patient satisfaction was higher when nurses, as opposed to doctors, provided first contact care for people wanting urgent attention. Patient satisfaction with chronic disease management was also found to higher with nurse-led care, although this was investigated in only one study. The reason for this difference is unclear and may relate to a number of factors. Nurses tended to have longer consultations than doctors, and patient satisfaction is higher with longer consultations (Freeman 2002). Nurses also tended to provide more information to patients than did doctors which might also have enhanced satisfaction. High satisfaction with nurse care did not, however, mean that patients inevitably preferred nurses to doctors. Patient preferences in most studies were mixed with some patients preferring to see nurses while others preferred to see doctors. Preference might partly relate to the nature of the presenting problem. Nurses may be preferred when the patient believes their problem to be ’minor’ or ’routine’ but doctors are preferred when the problem is thought to ’serious’ or ’difficult’ (Drury 1988). Productivity was lower when nurses, as opposed to doctors, provided first contact care for people wanting urgent attention. Nurses tended to have longer consultation lengths and higher rates of patient recall while achieving the same health outcomes as doctors. This might be a learning effect whereby nurse productivity would improve as nurses gained more experience in their role. However, two of the three studies which assessed productivity used experienced nurses (Kinnersley 2000; Venning 2000) and only one
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 11: research). Even so, relevant research proved difficult to identify and some papers may have been missed, particularly in the ’grey’ literature that we did not search. Publication bias seems unlikely as the clinical and research communities are interested equally in whether nurses outperform doctors or the reverse. The inclusion of only English and Dutch language publications risks excluding potentially relevant work. We did, however, screen the English abstracts of papers published in other languages and found none that appeared relevant. Research into doctor-nurse substitution in primary care appears primarily to have been conducted in Canada, the USA and the UK which are English-speaking countries. We restricted our meta-analyses to Fixed Effect (FE) models. The FE approach provides reliable estimates of the average effect (and confidence interval) across included studies, but findings cannot be generalised beyond these studies to the wider population of practitioners and practices. Therefore, where we have generalised beyond the studies in hand, this has been a qualitative judgement based on assessment of all the available evidence of which the metaanalysis is just one component. Most studies included only small numbers of nurses and very few considered the potential for variation in outcomes by practitioner. This may have led to over precision in the estimates of differences between doctors and nurses. In addition, studies intended to demonstrate the comparability of nurse and doctor care need to be powered to assess the equivalence, not difference, of outcomes. This was done in only one study (Lattimer 1998). A final concern is the narrow range of nurse roles that has been subjected to rigorous evaluation. Nurses in many countries provide a far wider range of care than is represented in the current research literature. Nursedoctor substitution in the management of patients with particular chronic diseases has been infrequently studied.
Nurse-doctor substitution has the potential to reduce the direct costs of care. Cost savings are, however, highly dependent on salary differentials between doctors and nurses and these may vary across locations and over time. In addition, savings on nurse salaries may be offset by nurses’ longer consultation length and increased rate of patient recall relative to doctors, leading to no overall savings on cost. Implications for research Cost, particularly societal cost, has not been well investigated despite the widely held view that nurse-led care will generate savings. Future studies of nurse-doctor substitution should give more attention to the financial aspects of care. Related to this is the question of what impact nurses have on doctor behaviour and workload. This has rarely been evaluated despite the widely held view that nurses can ’save’ doctors’ time. The methodological quality of studies is variable. Future studies should seek to maximise the numbers of practitioners (particularly nurses), rather than numbers of patients, in order to reduce the effect of any individual practitioner on outcomes. Studies also need to adopt methods of statistical analysis that account for variation in outcomes between practitioners, to avoid over precision and an inflated risk of type 1 errors (false positive results). Studies intended to demonstrate the comparability of nurse and doctor care need to be powered to assess the equivalence, not difference, of outcomes. A final concern is the narrow range of nurse roles that has been subjected to rigorous evaluation. Nurses in many countries provide a far wider range of care than is represented in the current research literature. Related to this is the question of what levels of training and experience are required by nurses working as doctor substitutes. The characteristics of nurses and doctors (numbers, training, experience) need to be reported more often and more consistently in studies in order to shed light on this issue.
AUTHORS’ CONCLUSIONS Implications for practice The findings suggest that appropriately trained nurses can produce as high quality care as primary care doctors and achieve as good health outcomes for patients. Indeed nurses providing first care for patients needing urgent attention tend to provide more health advice and achieve higher levels of patient satisfaction compared with doctors. Nurse-doctor substitution has the potential to reduce doctors’ workload. However this benefit will not be realised in practise if doctors continue to provide the types of care that have been transferred to nurses. Doctors’ workload may remain unchanged either because there was previously unmet need that nurses now fulfil or because nurses generate demand for care where previously there was none.
POTENTIAL CONFLICT OF INTEREST None known.
ACKNOWLEDGEMENTS We thank Steve Rose and Richard Heywood who advised on the search strategy and helped with the computerised searches; Michelle Sergison who helped design the review and assisted with early data extraction; and Shirley Halliwell who obtained the articles and helped to screen papers suitable for data extraction.
9
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 12: SOURCES OF SUPPORT External sources of support • Ministry of Health, Welfare and Sports NETHERLANDS • Department of Health UK Internal sources of support • Centre for Quality of Care Research (WOK), University of Nijmegen NETHERLANDS • National Primary Care Research and Development Centre (NPCRDC), University of Manchester UK
REFERENCES
References to studies included in this review
Chambers 1977 {published data only} Black DP, Riddle RJ, Sampson E. Pilot project: The family practice nurse in a Newfoundland rural area. Canadian Medical Association Journal 1976;114:945. Chambers LW, Bruce-Lockhart P, Black DP, Sampson E, Burke M. A controlled trail of the impact of the family practice nurse on volume, quality and cost of rural health services. Medical Care 1977; XV(12):971–81.
∗
evidence from a randomised controlled trial. BMJ 2000;320:1053– 7. South Wiltshire Out of Hours Project (SWOOP) Group. Nurse telephone triage in out of hours primary care: a pilot study. BMJ 1997; 314:198–9. Thompson F, George S, Lattimer V, Smith H, Moore M, Turnbull J, Mulle M, Thomas E, Bond H, Glasper A. Overnight calls in primary care: randomised controlled trial of management using nurse telephone consultation. BMJ 1999;319:1408. Lewis 1969 {published data only} ∗ Lewis CE, Resnik BA. Nurse clinics and progressive ambulatory patient care. New England Journal of Medicine 1969;1969(23):1236– 41. McIntosh 1997 {published data only} ∗ McIntosh MC, Leigh G, Baldwin NJ, Marmulak J. Reducing alcohol consumption. Comparing three brief methods in family practice. Canadian Family Physician 1997;43:1959–67. Moher 2001 {published data only} ∗ Moher M, Yudkin P, Wright L, Turner R, Fuller A, Schofield T, Mant D for the Assessment of Implementation Strategies (ASSIST) Trial Collaborative Group. Cluster randomised controlled trial to compare three methods of promoting secondary prevention of coronary heart disease in primary care. BMJ 2001;322:1338–42. Mundinger 2000 {published data only} ∗ Mundinger MO, Kane Rl, Lenz ER, Totten AM, Tsji WY, Cleary Pd, Friedewald WT, Siu AL, Shelanski ML. Primary care outcomes in patients treated by nurse practitioners or physicians. A randomized trial. JAMA 2000;283(1):59–68. Sox HC. Independent primary care practice by nurse practitioners. [editorial]. JAMA 2000;283(1):106–108. Myers 1997 {published data only} ∗ Myers PC, Lenci B, Sheldon MG. A nurse practitioner as the first point of contact for urgent medical problems in a general practice setting. Family Practice 1997;14(6):492–7. Shum 2000 {published data only} ∗ Shum C, Humphreys A, Wheeler D, Cochrane MA, Skoda S, Clement S. Nurse management of patients with minor illnesses in
10
Chambers 1978 {published data only} ∗ Chambers LW, West AE. St John’s Randomized trial of the family practice nurse: health outcomes of patients. International Journal of Epidemiology 1978;7(2):153–61. Flynn 1974 {published data only} Flynn BC. The effectiveness of nurse clinicans service delivery. American Journal of Public Health 1974;64(6):604–11. Gordon 1974 {published data only} ∗ Gordon DW. Health maintenance service: ambulatory patient care in the general medical clinic. Medical Care 1974;XII(8):648–58. Hemani 1999 {published data only} ∗ Hemani A, Rastegar DA, Hill C, Al-Ibrahim MS. A comparison of resource utilization in nurse practitioners and physicians. Effective Clinical Practice 1999;2(6):258–65. Kinnersley 2000 {published data only} ∗ Kinnersley P, Anderson E, Parry K, Clement J, Arcarhd L, Turton P, Stainthorpe A, Fraser A, Butler CB, Rogers C. Randomised controlled trial of nurse practitioner versus general practitioner care for patients requesting “same day” consultations in primary care. BMJ 2000;320:1043–8. Lattimer 1998 {published data only} ∗ Lattimer V, George S, Thompson F, Thomas E, Mulle M, Turnbull J, Smith H, Moore M, Bond H, Glasper A (the South Wiltshire Out of Hours Project (SWOOP) Group). Saftety and effectiveness of nurse telephone consultation in out of hours primary care: randomised controlled trial. BMJ 1998;317:1054–9. Lattimer V, Sassi F, George S, Moore M, Turnbull J, Mullee M. Cost analysis of nurse telephone consultation in out of hours primary care:
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 13: general practice: multicentre randomised controlled trial. BMJ 2000; 320:1038–43. Spitzer 1973 {published data only} ∗ Sackett DL, Spitzer WO, Gent M, Roberts RS (in collaboration with: Hay WI, Lefroy GM, Sweeny GP. Vandervlist I, Sibley JC, Chambers LW, Goldsmith CH. Macpherson AS, McAuley RG). The Burlington randomized trial of the nurse practitioner: health oucomes of patients. Annals of Internal Medicine 1974;80:137–42. Spitzer WO, Kergin DJ. Nurse practitioners in primary care. I. The McMaster University educational program. Canadian Medical Association Journal 1973;108:991–5. Spitzer WO, Roberts RS, Delmore T. Nurse practitioners in primary care. V. Development of the utilization and financial index to measure effects of their deployment. Canadian Medical Association Journal 1976;114:1099–1102. Spitzer WO, Roberts RS, Delmore T. Nurse practitioners in primary care. VI. Assessment of their deployment with the utilization and finacial index. Canadian Medical Association Journal 1976;114:1103– 8. Sweeny GP, Ian Hay WI. The Burlington experience: a study of nurse practitioners in family practice. Canadian Family Physician 1973;19: 101–10. Stein 1974 {published data only} ∗ Stein GH. The use of a nurse practitioner in the management of patients with diabetes mellitus. Medical Care 1974;XII(10):885–90. Venning 2000 {published data only} ∗ Venning P, Durie A, Roland M, Roberts C, Leese B. Randomised controlled trail comparing cost effectiveness of general practitioners and nurse practitioners in primary care. BMJ 2000;320:1048–53.
Spitzer WO, Kergin DJ. Nurse practitioners in primary care. I. The McMaster University educational program. Canadian Medical Association Journal 1973;108:991–5. Spitzer WO, Roberts RS, Delmore T. Nurse practitioners in primary care. V. Development of the utilization and financial index to measure effects of their deployment. Canadian Medical Association Journal 1976;114:1099–1102. Spitzer WO, Roberts RS, Delmore T. Nurse practitioners in primary care. VI. Assessment of their deployment with the utilization and finacial index. Canadian Medical Association Journal 1976;114:1103– 8. Campbell 1998 ∗ Campbell NC, Thain J, Deans HG, Ritchie LD, Rawles JM, Squair JL. Secondary prevention clinics for coronary heart disease: randomized trial of effect on health. BMJ 1998;316:1434–7. Cargill 1991 ∗ Cargill VA, Conti M, Neuhauser D, McClish D. Improving the effectiveness of screening for colorectal cancer by involving nurse clinicians. Medical Care 1991;29(1):1–5. Cherkin 1996 ∗ Cherkin DC, Deyo RA, Street JH, Hunt M, Barlow W. Pitfalls of patient education. Limited success of a program for back pain in primary care. Spine 1996;21(3):345–55. Cupples 1994 Cupples ME, McKnight A. Five year follow up of patients at high cardiovascular risk who took part in randomized controlled trial of health promotion. BMJ 1999;319:687–8. Cupples ME, McKnight A. Randomised controlled trial of health promotion in general practice for patients at high cardiovascular risk. BMJ 1994;309:993–6.
∗
References to studies excluded from this review
Bakx 1997 ∗ Bakx JC, Stafleu A, van Staveren WA, van den Hoogen HJM, van Weel C. Long-term effect of nutritional counseling: a study in family medicine. American Journal of Clinical Nutrition 1997;65(suppl): 1946S–50S. van Ree, JW, van Gerwn W, van den Hoogen H. [Interventie bij een verhoogd risico op hart-en vaatziekten. 1.Resultaten op langere termijn van interventie bij hypercholesterolemie, adipositas en roken]. Huisarts en Wetenschap 1985;28:21–4. van Ree, JW, van Gerwn W, van den Hoogen H. [Interventie bij een verhoogd risico op hart-en vaatziekten. 2. Resultaten op langere termijn bij de behandeling van hypertensie]. Huisarts en Wetenschap 1985;28:54-8, 77. Batchelor 1975 ∗ Batchelor GM, Spitzer WO, Comley AE, Anderson GD. Nurse practitioners in primary care IV. Impact of an interdisciplinary team on attitudes of a rural population. Canadian Medical Association Journal 1975;112:1415–20. Chenoy NC, Spitzer WO, Anderson GD. Nurse practitioners in primary care II. Prior attitudes of a rural population. Canadian Medical Association Journal 1973;108:998–1003.
Cupples ME, McKnight A, O’Neill C, Normand C. The effect of personal health education on the quality of life of patients with angina in general practice. Health education journal 1996;55:75–83. Fall 1997 ∗ Fall M, Walters S, Read S, Deverill M, Lutman M, Milner P, Rodgers R. An evaluation of a nurse-led ear care service in primary care: benefits and costs. British Journal of General Practice 1997;47:699– 703. Family Heart 1994 a Family Heart Study Group. Its design and method, and prevalence of cardiovascular risk factors. British Journal of General Practice 1994; 44:62–7.
∗ Family Heart Study Group. Randomised controlled trial evaluating cardiovascular screening and intervention in general practice: principal results of British family heart study. BMJ 1994;308:313–20.
Wonderling D, McDermott, Buxton M, knmonth A, Pyke S, Thompson S, Wood D. Costs and cost effectiveness of cardiovascular screening and intervention: the British family heart study. BMJ 1996;312:1269–73. Fullard 1987 Fullard E. Extending the roles of practice nurses and facilitators in preventing heart disease. The Practitioner 1987;231:1283–6.
11
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 14: Fullard E, Fowle G, Gray M. Promoting prevention in primary care: controlled trial of low technology, low cost approach. BMJ 1987; 294:1080–2. Goldberg 1991 ∗ Goldberg HI, Mullen M, Ries Rk, Psatty Bm, Ruch BP. Alcohol counseling in a general medicine clinic. A randomized controlled trial of strategies to improve referral and show rates. Medical Care 1991; 29(7 Supplement):JS49–JS56..
∗
and total serum cholesterol level. Archives of Family Medicine 1997; 6:129–34. Ridsdale 1996 a Ridsdale L, Robins D, Cryer C, Williams H, and the Epilepsy Care Evaluation Group. Feasibility and effects of nurse run clinics for patients with epilepsy in general practice: randomized controlled trial. BMJ 1997;314:120–2. Ridsdale L, Robins D, Fitzgerald A, Jeffery S, McGee L (Epilepsy Care Evaluation Group). Epilepsy in general practice: patients’ psychological symptoms and their perception of stigma. British Journal of General Practice 1996;46:365–6. Ridsdale L, Robins D, Fitzgerald A, Jeffery S, McGee L (Epilepsy Care Evaluation Group). Epilepsy monitoring and advice recorde: general practitioners’ views, current practice and patients’ preferences. British Journal of General Practice 1996;46:11–4.
∗
Jamrozik 1984 ∗ Jamrozik K, Vessey M, Fowler G, Wald N, Parker G, van Vunakis H. Controlled trial of three different antismoking interventions in general practice. British Medical Journal 1984;288:1499–1503. Klerman 1987 ∗ Klerman GL, Budman S, Berwick D, Weissman MM, DamicoWhite J, Demby A, Feldstein M. Efficacy of a brief psychosocial intervention for symptoms of stress and distress among patients in primary care. Medical Care 1987;25(11):1078–88. Mann 1998 ∗ Mann AH, Blizard R, Murray J, Smith JA, Botega N, Macdonanld E, Wilkinson G. An evaluation of practice nurses working with general practitioners to treat people with depression. British Journal of General Practice 1998;48:875–9. Margolis 1996 ∗ Margolis PA, Lannon CM, Stevens R, Harlan C, Bordley WC, Carey T, Leininger L, Keyes LL, Earp JaL. Linking clinical and public helath approaches to improve access to health care for socially disadvantaged mothers and children. Archives of Pediatric Adolescent Medicine 1996;150:815–21. Muir J 1995 Langham S, Thorogood M, Normand C, Muir J, Jones L, Fowler G. Costs and cost effectiveness of health checks conducted by nurses in primary care: the OXCHECK study. BMJ 1996;312:265–8. Muir J, Lancaster T, Jones L, Yudkin P (for the Imperical Cancer Research Fund OXCHECK Study Group). Effectiveness of health checks conducted by nurses in primary care: final results of the OXCHECK study. BMJ 1995;310:1099–1104.
∗
Scambler A, Scambler G, Ridsdale L, Robins D. Towards an evaluation of the effectiveness of an epilepsy nurse in primary care. Seizure 1996;5:255–8. Robson 1989 ∗ Robson J, Boomla K, Fitzpatrick S, Jewell AJ, Taylor J, Self J, Colyer M. Using nurses for preventive activities with compter assisted follw up: a randomized controlled trial. BMJ 1989;298:433–6. Sanders 1989 ∗ Sanders D, Fowler G, Mant D, Fuller A, Jones L, Marzillier J. Randomized controlled trial of anti-smoking advice by nurses in general practice. Journal of the Royal College of General Practitioners 1989;39:273–6. Sanders D, Peveler R, Mant D, Fowler G. Predictors of successful smoking cessation following advice from nurses in general practice. Addiction 1993;88:1699–1705. Sharp 1996 ∗ Sharp DJ, Peters TJ, Bartholemew J, Shaw A. Breast screening: a randomized controlled trial in UK general practice of three interventions designed to increase uptake. Journal of Epidemiology and Community Health 1996;50:72–6. Thompson 1982 ∗ Thompson S, Basden P, Howell LJ. Evaluation of initial implementation of an organized adult health program employing family nurse practitioners. Medical Care 1982;XX(11):1109–27. Vetter 1984 Vetter NJ, Jones DA, Victor CR. A health visitor affects the problems others do not reach. The Lancet 1986;2:30–2. Vetter NJ, Jones DA, Victor CR. Effect of health visitors working with elderly patients in general practice: a randomized controlled trial. British Medical Journal 1984;288:369–72.
∗
Muir J, Mant D, Jones L, Yudkin P (the Imperical Cancer Research Fund OXCHECK Study Group). Effectivenees of health checks conducted by nurses in primary care: results of the OXCHECK study after one year. BMJ 1994;308:308–12. O’Neill C. The OXCHECK study: a nursing perspective. Health reports 1994;6(1):160–5. Wonderling D, Langham S, Buxton M, Normand C, McDermott C. What can be concluded from the OXCHECK and british family heart studies: commentary on cost effectiveness analyses. BMJ 1996; 312:1274–8. Mynors-Wallis 1997 ∗ Mynors-Wallis L, Davies I, Gray A, Barbour F, Gath D. A randomized controlled trial and cost analysis of problem-solving treatment for emotional disorders given by community nurses in primary care. British Journal of Psychiatry 1997;170:113–9. Pine 1997 ∗ Pine DA, Madlon-Kay DJ. Sauser M. Effectiveness of a nurse-based intervention in a community practice on patients’ dietary fat intake
Vetter 1992 ∗ Vetter NJ, Lewis PA, Ford D. Can health visitors prevent fractures in elderly people?. BMJ 1992;204:888–90. Wilkinson 1993 ∗ Wilkinson G, Allen P, Marshall E, Walker J, Browne W, Mann AH. The role of the practice nurse in the management of depression in general practice: treatment adherence to antidepressant medication. Psychological Medicine 1993;23:229–37.
12
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 15: Woollard 1995 ∗ Woollard J, Beilling L, Lord T, Puddey I, MacAdam D, Rouse I. A controlled trial of nurse counseling on lifestyle change for hypertensives treated in general practice: preliminary results. Clinical and Experimental Pharmacology and Physiology 1995;22:466–8.
Freeman 2002 Freeman GK, Horder JP, Howie JGR, Hungin AP, Hill AP, Shah NC, Wilson A. Evolving general practice consultation in Britain: issues of length and context. BMJ 2002;324:880–2. Horrocks 2002 Horrocks, S, Anderson, E, Salisbury, C. Systematic review of whether nurse practitioners working in primary care can provide equivalent care to doctors. BMJ 2002;324:819–23. Jenkins-Clarke 1998 Jenkins-Clarke S, Carr-Hill R, Dixon P. Teams and Seams: skill mix in primary care. Journal of Advanced Nursing 1998;28:1120–6. Kirkman 1994 Kirkman MS, Weinberger M, Landsman PS, Sams GP, Shortlife EA, Simel DL, et al. A telephone-delivered intervention for patients with NIDDM. Diabetes Care 1994;17:840–6. Laurant 2004 Laurant, M, Hermens, R, Braspenning, J, Grol, R, SIbbald, B. Nurse practitioners do not reduce the workload of GPs - a randomized controlled trial. BMJ 2004;328:927–30. Maas 2004 Maas, CMJ, Hox, JJ. Robustness issues in multilevel regression analysis. Statistica Neerlandica 2004;58(2):127–37. Muir 1995 .Muir J, Lancaster T, Jones L, Yudkin P, for the OXCHECK Study Group. Effectiveness of health checks conducted by nurses in primary care: final results of the OXCHECK study. BMJ 1995;310:1099– 104. Petitti 2001 Petitti, D. Approaches to heterogeneity in meta-analysis. Statistics in Medicine 2001;30:3625–33. Richardson 1998 Richardson G, Maynard A, Cullum N, Kindig, D. Skill mix changes: substitution or service development?. Health Policy 1998;45(2):119– 32. Richardson 1999 Richardson MSC. Identifying, evaluating and implementing costeffective skill mix. J Nurse Manage 1999;5:265–270. Whitecross 1999 Whitecross L. Collaboration between GPs and nurse practitioners. The overseas experience and lessons for Australia. Australian Family Physician 1999;28:349–53.
∗
Additional references
Alderson 2004 Alderson P, Green S, Higgins JPT, editors. Cochrane Reviewers’ Handbook 4.2.2 [updated March 2004]. In: The Cochrane Library [database on CDROM]. The Cochrane Collaboration. Oxford: Update Software; 2005, issue 1. Aubert 1998 Aubert RE, Herman WH, Waters J, Moore W, Sutton D, Peterson BL, et al. Nurse case management to improve glycaemic control in diabetic patients in a health maintenance organization. Annals of Internal Medicine 1998;129:605–12. Bailey 1987 Bailey, KR. Inter-study differences: how should the influence the interpretation and analysis of the results?. Statistics in Medicine 1987; 6(3):351–60. Biggerstaff 1997 Biggerstaff, BJ, Tweedie, RL. Incorporating variability in estimates of heterogeneity in the random effects model in meta-analysis. Statistics in Medicine 1997;16:753–68. Brockwell 2001 Brockwell, SE, Gordon, IR. A comparison of statistical methods for meta-analysis. Statistics in Medicine 2001;20:825–40. Brown 1995 Brown, SA, Grimes, DE. A meta-analysis of nurse practitioners and nurse midwives in primary care. Nursing Research 1995;44:332–9. Charlton 1991 Charlton I, Charlton G, Broomfield J, Mullee MA. Audit of the effect of a nurse run asthma clinic on workload and patient morbidity in a general practice. British Journal of General Practice 1991;41(347): 227–31. Drury 1988 Drury M, Greenfield S, Stillwell B, Hull FM. A nurse practitioner in general practice: patients perceptions and expectations. Journal of the Royal College of General Practitioners 1988;38:503–5. Family HSG 1994 Family Heart Study Group. Randomised controlled trial evaluating cardiovascular screening and intervention in general practice: principal results of British Family Heart Study. BMJ 1994;308:313–20.
Indicates the major publication for the study
TABLES
Characteristics of included studies
Study Methods Chambers 1977 CBA
13
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 16: Characteristics of included studies (Continued )
Participants 2313 patients, all ages, 52% male 1 nurse Unknown number of doctors Intervention: two villages allocated to nurse-led care Control: neighbouring villages allocated to doctor-led care Process of care: standards of care Resource utilisation: direct costs Nurse title: practice nurse Nurse role: First contact care and ongoing primary care Study period: 12 months Allocation concealment Study Methods Participants D – Not used Chambers 1978 RCT 868 patients, all ages, 34% male 1 nurse 1 doctor Intervention: families allocated to nurse-led primary care Control: families allocated to doctor-led primary care Patient outcomes: health status Nurse title: practice nurse Nurse role: First contact and ongoing primary care Study period: 12 months Allocation concealment Study Methods Participants B – Unclear Flynn 1974 RCT 60 patients, age unknown, gender unknown 4 nurses Unknown number of doctors Intervention: patients allocated to nurse-led primary care Control: patients allocated to doctor-led primary care Patient outcomes: compliance with medication and diet; knowledge Process of care: suggested lifestyle changes Resource utilisation: tests and investigations; use of other health services Notes Nurse title: nurse clinician Nurse role: First contact and ongoing primary care Study period: 6-12 months Allocation concealment Study Methods Participants B – Unclear Gordon 1974 CBA 169 patients, all ages, 38% female.
14
Interventions Outcomes Notes
Interventions Outcomes Notes
Interventions Outcomes
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 17: Characteristics of included studies (Continued )
Interventions Outcomes Unknown number of nurses and doctors Intervention: patients allocated to nurse-led primary care Control: patients allocated to doctor-led primary care Patient outcomes: health status; satisfaction; compliance with medication and follow-up attendance. Process of care: lapses in care Nurse title: nurse clinician Nurse role: First contact and ongoing primary care Study period: 12 months Allocation concealment Study Methods Participants D – Not used Hemani 1999 RCT 450 patients, mean age 61 years, 98% male 9 nurses 45 doctors Intervention: patients allocated to nurse-led primary care Control 1: patients allocated to trainee doctors (2nd ,3rd year residents) Control 2: patients allocated to fully trained doctors (attending physicians) Resource utilisation: consultation rate; tests;, use of other services-hopsital admission, emergency room visits, specialty visits Nurse role: First contact and ongoing primary care Study period: 12 months Allocation concealment Study Methods Participants A – Adequate Kinnersley 2000 RCT 1465 patients, all ages, 40% male 10 nurses Unknown number of doctors in 10 practices Intervention: patients allocated to nurse Control: patients allocated to doctor Patient outcomes: health status; satisfaction; provider preference Process of care: provision of information Resource utilisation: length of consultation; return visits; prescriptions; investigations; use other services referral Nurse title: nurse practitioner Nurse role: First contact care for patients with urgent problems Study period: 2-4 weeks Allocation concealment Study Methods A – Adequate Lattimer 1998 RCT
15
Notes
Interventions
Outcomes Notes
Interventions Outcomes
Notes
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 18: Characteristics of included studies (Continued )
Participants 10134 patients, all ages, 48% male 6 nurses 55 doctors Intervention: incoming phone calls on randomly selected days were allocated to nurse telephone consultation. Control: incoming phone calls on other days were answered by a receptionist who passed the message to a doctor Patient outcomes: mortality Resource utilisation: physician workload, use of other services - hospital referral and admission, emergency room visits, direct costs. Notes Nurse title: not clear Nurse role: First contact care for patients with urgent problems out-of-hours Study period: 3-7 days Allocation concealment Study Methods Participants Interventions Outcomes A – Adequate Lewis 1969 RCT 66 patients, 16+ years, 12% male Unknown number of nurses and doctors Intervention: patients allocated to nurse-led care Control: patients allocated to doctor-led care Patient outcomes: health status; provider preference; compliance with follow-up attendance Resource utilisation - direct costs Nurse title: not clear Nurse role: Ongoing primary care for patients with stable chronic disease Study period: 12 months Allocation concealment Study Methods Participants B – Unclear McIntosh 1997 RCT 119 patients, mean age 31.5 years, 50% male 1 nurse 1 doctor Intervention: patients allocated to nurse Control: patients allocated to doctor who provided identical treatment Patient outcomes: alcohol consumption Nurse title: nurse practitioner Nurse role: Counselling to problem drinkers Study period: 12 months Allocation concealment Study Methods Participants B – Unclear Moher 2001 RCT 1347 patients, mean age 66 years, 69% male
16
Interventions
Outcomes
Notes
Interventions Outcomes Notes
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 19: Characteristics of included studies (Continued )
Unknown number of nurses and doctors in 21 practices Interventions Outcomes Intervention: patients allocated to nurse-led follow-up Control: patients allocated to doctor-led follow-up Patient outcomes: cardiovascular risk factors Process of care: adherence to guidelines. Resource utilisation: prescriptions Notes Nurse title: practice nurse Nurse role: Ongoing primary care for patients with coronary heart disease Study period: 18 months Allocation concealment Study Methods Participants B – Unclear Mundinger 2000 RCT 1316 patients, mean age 44.5 years, 25.5% male 7 nurses 17 doctors Intervention: patients allocated to nurse-led care Control: patients allocated to doctor-led care Patient outcomes: health status; satisfaction Resource utilisation: consultation rate; use of other services -hospital admissions, emergency room visits, specialty visits Notes Nurse title: nurse practitioner Nurse role: First contact and ongoing primary care Study period: 6 months Allocation concealment Study Methods Participants B – Unclear Myers 1997 CBA 1000 patient contacts, mean age 35.5 years, 40% male 2 nurses 6 doctors Intervention: patients choosing nurse Control: patients choosing doctor Resource utilisation:prescriptions; use of other services - referral Nurse title: nurse practitioner Nurse role: First contact care for patients with urgent problems Study period: 14 days Allocation concealment Study Methods Participants D – Not used Shum 2000 RCT 1815 patients, mean age 27.5 years, 40% male 5 nurses
17
Interventions Outcomes
Interventions Outcomes Notes
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 20: Characteristics of included studies (Continued )
Interventions Outcomes 19 doctors Intervention: patients allocated to nurse Control: patients allocated to doctor Patient outcomes: health status; satisfaction; provider preference Process of care: provision of information Resource utilisation: length of consultation; return visits; prescriptions; use other services - emergency room visits, use of out-of-hour services Nurse title: practice nurse Nurse role: First contact care for patients with urgent problems Study period: 2 weeks Allocation concealment Study Methods Participants A – Adequate Spitzer 1973 RCT 4325 patients, all ages, 42.5% male 2 nurses 2 doctors Intervention: families allocated to nurse Control: families allocated to doctor Patient outcomes: health status; satisfaction, provider preference Process of care: Standards of care Notes Resource utilisation: direct costs Nurse title: nurse practitioner Nurse role: First contact and ongoing primary care Study period: 12 months Allocation concealment Study Methods Participants B – Unclear Stein 1974 RCT 23 patients, mean age 56 years, 0% male 1 nurse Unknown number of doctors Intervention: patients allocated to nurse-led care Control: patients allocated to doctor-led care Patient outcomes: health status; mortality; knowledge Resource utilisation: consultation rate; prescriptions Notes Nurse title: nurse practitioner Nurse role: Ongoing care for patients with diabetes mellitus Study period: 6 months Allocation concealment Study Methods C – Inadequate Venning 2000 RCT
18
Notes
Interventions Outcomes
Interventions Outcomes
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 21: Participants
Interventions Outcomes
1316 patients, all ages, 42% male 20 nurses Unknown number of doctors Intervention: patients allocated to nurse Control: patients allocated to doctor Patient outcomes: health status; satisfaction; compliance with follow-up attendance; enablement Process of care: examinations Resource utilisations: length of consultation; return visits; prescriptions; investigations; use of other serviceshospital referral, direct costs
Notes
Nurse title: nurse practitioner Nurse role: First contact care for patients with urgent problems Study period: 2 weeks
Allocation concealment
A – Adequate
Participants: Where the mean age or proportion of males in the study group as a whole was not reported, we have estimated these values by averaging the figures given for each intervention group. Outcomes: Only usable outcomes (i.e. amenable to statistical analysis) are listed Allocation Concealment: A - Adequate; B - Unclear; C- Inadequate; D - Not used.
Characteristics of excluded studies
Study Bakx 1997 Batchelor 1975 Campbell 1998 Cargill 1991 Cherkin 1996 Cupples 1994 Fall 1997 Fullard 1987 Goldberg 1991 Jamrozik 1984 Klerman 1987 Mann 1998 Margolis 1996 Muir J 1995 Mynors-Wallis 1997 Pine 1997 Ridsdale 1996 a Robson 1989 Sanders 1989 Sharp 1996 Reason for exclusion Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements This paper describes to two studies, both are excluded because nurses work as supplements. Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements
19
Family Heart 1994 a Nurses working as supplements
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 22: Characteristics of excluded studies (Continued )
Thompson 1982 Vetter 1984 Vetter 1992 Wilkinson 1993 Woollard 1995 Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements Nurses working as supplements
ADDITIONAL TABLES
Table 01. Methodological Quality of controlled before and after studies
Unit Analysis Error
Study (no) Chambers 1977
Power
80% follow- Comparaup bility Professional - Not clear; Patients Done Professional - Not clear; Patients Done Professional - Not clear; Patients Done Done
Baseline Assessment Done
Blinded Assessment Done
Reliable Outcomes Not clear
Contamination Not clear
Not clear Yes
Gordon 1974
Not clear No
Done
Not clear
Resource Not clear - Done; Patient - Not done Done Not clear
Not clear
Myers 1997
Not clear No
Done
Not clear
Not done
Table 02. Methodological quality of randomized controlled trials
Unit Analysis Error Yes
Study (no) Chambers 1978
Power Not done
80% follow- Concealment up Professional - Not clear; Patient - not done Professional - Not clear; Patient done Professional - Not clear; Patient done Not clear
Baseline Assessment Done
Blinded Assessment Not clear
Reliable Outcomes Not clear
Contamination Not clear
Flynn 1974
Not clear
No
Not clear
Not clear
Not done
Not done
Not clear
Hemani 1999
Not done
No
Done
Not clear
Done
Done
Not clear
Kinnersley 2000
Done
No
Professional Done - Not clear; Patient - Not
Not clear
Not done
Done
Not clear
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
20
Page 23: Table 02. Methodological quality of randomized controlled trials (Continued )
Unit Analysis Error
Study (no)
Power
80% follow- Concealment up done
Baseline Assessment
Blinded Assessment
Reliable Outcomes
Contamination
Lattimer 1998
Done
No
Professional - Not clear; Patient done Professional - Not clear; Patient done Professional - Not clear; Patient done Professional - Not clear; Patient done
Done
Not clear
Done
Done
Not clear
Lewis 1967
Not clear
No
Not clear
Not clear
Done
Not clear
Not clear
McIntosh 1997
Done
No
Not clear
Done
Not done
Done
Not clear
Moher 2001
Not clear
No
Not clear
Done
Done
Not clear
Done
Mundinger 2000
Done
No
Professional Not clear - Not clear; Patient - Not done Professional - Not clear; Patient Done Done
Done
Resource Done; Patient - Not clear Not done
Resource Not clear Done; Patient - Not clear Done Not clear
Shum 2000
Done
No
Not clear
Spitzer 1974
Not clear
Yes
Professional Not clear - Not clear; patient - Not clear Professional - Not clear; Patient Done Not done
Done
Mortality - Done; Others - Not clear Not clear
Not clear Mortality - Done; Others - Not clear Not clear Not clear
Stein 1974
Not clear
No
Not clear
Venning 2000
Not clear
No
Professional Done - Not clear; Patient - Not done
Not clear
Not done
Health status Not clear - done; Others - Not clear
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
21
Page 24: Table 03. Patient outcomes
Study (no) Chambers 1978 Nurse Role Health status First contact and ongoing care Satisfaction Health status: -Physical function: nurse better -Emotional function: no difference -Social function: no difference Medication: no difference Diet: no difference Patients’ knowledge: - Exercise: nurse significantly better - Disease complications: no difference - Diet: no difference Compliance Other
Flynn 1974
First contact and ongoing care
Gordon 1974
First contact and ongoing care
Health Status: no difference
Subjective perceptions of clinical care (10 dimensions): no difference Satisfaction (3 dimensions) no difference in overall satisfaction, but nurse worse on 1 dimension Would recommend provider to others: no difference
Medication: no difference Kept appointment: no difference
Mundinger 2000
First contact and ongoing care
Number of Health Complaints no difference Health Status (10 dimensions): No difference Objective Measures of patient health: - Asthma - peak flow: no difference - Diabetes - blood sugar: no difference - Hypertension: systolic pressure: no difference Hypertension: disatolic pressure: nurse better Health status: -Physical function (3 indicators): no difference -Emotional function: no difference -Social function: no difference Mortality: no difference
Spitzer 1973
First contact and ongoing care
Satisfaction: no difference
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
22
Page 25: Table 03. Patient outcomes (Continued )
Study (no)
Nurse Role
Health status
Satisfaction
Compliance
Other
Kinnersley
First contact care for
Health status: -resolution of
Satisfaction: - child
23
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 26: Table 03. Patient outcomes (Continued )
Study (no) 2000 Nurse Role urgent problems Health status symptoms: no difference resolution of concerns: no difference Mortality: no difference Health status: no difference Satisfaction: -general: nurse significantly better -professional care: nurse significantly better -relationship to provider: no difference -adequacy of time: nurse significantly better -explanation helpful: no difference -advice helpful: no difference Provider preference: patients preferred nurse significantly more often Satisfaction: Adults -General: nurse significantly better -Communication: nurse significantly better -Distress relief: nurse significantly better -Professional care: nurse significantly better Children -General: nurse significantly better -Communication with parent: no difference -Communication with child: nurse
24
Satisfaction care: nurse better -adult care: no difference Provider preference: no difference
Compliance
Other
Latimer 1998 Shum 2000
First contact care for urgent problems First contact care for urgent problems
Venning 2000
First contact care for urgent problems
Health status: no difference Objective measures of patient health: -Asthma - peak flow: no difference -Diabetes - blood sugar: no difference -Hypertension - systolic blood pressure: no difference -Hypertension diastolic blood pressure: nurse significantly better
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 27: Table 03. Patient outcomes (Continued )
Study (no) Nurse Role Health status Satisfaction significantly better -Distress relief: nurse significantly better -Adherence intent: no difference Lewis 1967 Management of patients with chronic conditions Management of patients with chronic conditions Management of patients with chronic conditions Health status: resolution of symptoms: nurse better. Health status: -reduction in alcohol consumption: no difference Objective measurement health status: -Blood pressure: no difference - Cholesterol: No difference - Not Smoking: No Difference Objective measurement health status: -Blood sugar: no difference - Weight: no difference Mortality: no difference Knowledge: nurse better Provider preference: nurse better Kept appointment: no difference Compliance Other
McIntosh 1997
Moher 2001
Stein 1974
Management of patients with chronic conditions
Table 04. Process of Care Outcome
Study (No) Chambers, 1977 Flynn, 1974 Nurse Role First contact and ongoing care Provider Care Adequate care: - clinical assessment: no difference - drug treatment: no difference Recommendations on: Ordered diet: - diabetic: no difference - low salt: nurse significantly higher frequency - low calorie: no difference - bland: no difference
25
First contact and ongoing care
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 28: Table 04. Process of Care Outcome
Study (No) Nurse Role
(Continued )
Provider Care Exercise: - Increase activities: nurse significantly higher frequency
Gordon, 1974
First contact and ongoing care
Lapses in care: no difference Subgroup: - stable patients: no difference - unstable patients: nurse significantly fewer lapses Provision of information: - Cause of illness: nurse significantly more - Relief of symptoms: nurse significantly more - Duration of illness: nurse significantly more - Reduce recurrence: nurse significantly more - Action if problem persists: no difference Provision of infomation: - Self-medication: nurse significantly more - Self-management: nurse significantly more Adequate treatment: - Drug treatment: no difference Management of episodes: no difference Examinations: no difference Adequate assessment: - clinical assessment: no difference - blood pressure: no difference - cholesterol: no difference - smoking status: no difference
Kinnersley, 2000
First contact and ongoing care
Shum, 2000
First contact and ongoing care
Spitzer, 1973 Venning, 2000 Moher, 2001
First contact and ongoing care
First contact and ongoing care Management of patietns with chronic conditions
Table 05. Resource Utilisation Outcomes
Study (no) Flynn, 1974 Nurse role First contact and ongoing care Consultations Presc’n. & Invest’n. Investigations & tests (11 indicators): -nurse significantly more tests for 4 indicators (electrocardiogram; bacteriology; urinalysis; minor X-ray); the remainder showed no difference. Consultation rate: No difference compared to qualified doctors; Nurse significantly more visits compared to trainee doctors Lab tests: (6 indicators) no difference Other services Nurse-led care was associated with a significantly higher use of other services
Hemani, 1999
First contact and ongoing care
Hospital admission: no difference Emergency room visits: no difference Specialty visits: no difference
26
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 29: Table 05. Resource Utilisation Outcomes
Study (no) Mundinger, 2000 Nurse role First contact and ongoing care
(Continued )
Presc’n. & Invest’n. Other services Hospital admissions: no difference Emergency room visits: no difference Speciality visits: no difference Prescriptions:No difference Investigation ordered: no difference Referral to hospital: no difference
Consultations Consultation rate: no difference
Kinnersley, 2000
First contact care for urgent Consultation length: Nurse problems significantly longer Return Visit: Recommended: No difference Re-attend for same problem: No difference First contact care for urgent Impact on doctors’ workload: problems Telephone advice from doctor: significantly fewer with nurse led care Surgery visits: Significanlty fewer with nurse-led care Home visits: Significanlty fewer with nurse-led care First contact care for urgent problems First contact care for urgent Consultation length: Nurse problems significantly longer Return visit: no difference First contact care for urgent Consultation length: Nurse problems significantly longer Return visit: All visits: nurse significantly more Asked to return: Nurse significantly more Management of patients with chronic conditions
Lattimer, 1998
Hospital admission within 24 hours: no difference Hospital admission with 3 days: no difference Emergency room visits: no difference Referred to hospital emergency room: no difference Prescriptions: Nurse signifciantly less Referral: no difference
Myers, 1997 Shum, 2000
Prescriptions: No difference Emergency room visit: no difference Out-of-hours-calls: no difference Prescriptions: All: no difference Antibiotics: no difference Investigations: Nurse significantly more Referral to hospital: no difference
Venning, 2000
Moher, 2001
Prescriptions: Antihypertensives: no difference Lipid lowering:no difference Antiplatelet:no difference Consultation rate: no difference Prescriptions: Changed medication: no difference
Stein, 1974
Management of patients with chronic conditions
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
27
Page 30: Table 06. Cost Outcomes
Study (no) Chambers, 1977 Nurse role Cost outcomes First contact and ongoing primary care Notes Direct cost per 1000 patients per year: nurses - increase of 26% from $68130 to $85.690 doctors - increase of 21% from $93190 to $112.730
Lattimer, 1998
First contact for patients with urgent problems out-of-hours
Annual direct cost nurse-led service - £81.237 more than doctor-led service Savings: generated in reduced hospital and primary care utilisation £94.422 Net reduction in costs with nurseled service £3728 - £12.3824 (determined by sensitivity analysis) Average cost per patient per year nurses - $297.01 doctors - $285.67 Spitzer reported an overall reduction in practice costs following the introduction of nurse practitioners but this finding was based on observational before-and-after data. Data obtained from the related randomised controlled trial (reported above) did not support this finding.
Spitzer, 1973
First contact and ongoing primary care
Venning, 2000
First contact care for patients with urgent problems
Total direct cost per consultation: Nurses - mean £18.11 (range £0.66 - £297.1) Doctors - mean £20.70 (range £0.78 - £300.6) not significantly different Total direct cost per year: nurses $3.251 doctors - $4.199 Average cost per patient per year: nurses - $98.51 doctors - $127.24
Lewis, 1967
Ongoing care for patients with stable chronic diseases
ANALYSES
Comparison 01. Doctor-Nurse substitution study results
Outcome title 01 Physical function (better vs not better) 02 Patient satisfaction 03 Scheduled return visits 04 Prescription ordered 05 Hospital referral No. of studies 3 3 3 3 3 No. of participants 3211 3611 4022 4212 17152 Statistical method Relative Risk (Fixed) 95% CI Standardised Mean Difference (Fixed) 95% CI Relative Risk (Fixed) 95% CI Relative Risk (Fixed) 95% CI Relative Risk (Fixed) 95% CI Effect size 1.02 [0.98, 1.05] 0.28 [0.21, 0.34] 1.34 [1.20, 1.49] 1.00 [0.96, 1.05] 0.79 [0.58, 1.07]
28
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 31: 06 Attendance at Accident&Emergency 07 Hospital admission
3 3
17140 15860
Relative Risk (Fixed) 95% CI Relative Risk (Fixed) 95% CI
1.04 [0.94, 1.15] 1.17 [1.04, 1.31]
INDEX TERMS Medical Subject Headings (MeSH) Family Practice [economics; ∗ organization & administration]; Health Services Needs and Demand [economics; ∗ organization & administration]; Nurse Practitioners [organization & administration]; Nurses [∗ organization & administration]; Personnel Delegation [∗ organization & administration]; Primary Health Care [economics; ∗ organization & administration] MeSH check words Humans
COVER SHEET Title Authors Contribution of author(s) Substitution of doctors by nurses in primary care Laurant M, Reeves D, Hermens R, Braspenning J, Grol R, Sibbald B Miranda Laurant and Bonnie Sibbald wrote the protocol, performed the searching, read the abstracts and retrieved and appraised relevant evidence. Two researchers (ML, BS) extracted the necessary data using a checklist developed by EPOC, modified and amended for the purposes of this review. David Reeves conducted the meta-analyses. Rosella Hermens, Jozé Braspenning, and Richard Grol acted as supervisors to these processes. Miranda Laurant wrote the first draft of the review. All authors were involved in the discussion. 1998/4 / 24 February 2005 16 July 2004 Information not supplied by author Information not supplied by author Information not supplied by author Information not supplied by author Information not supplied by author Dr Miranda Laurant Centre for Quality of Care Research University of Nijmegen (229 HSV/WOK) PO Box 9101 6500 HB Nijmegen 6500 HB NETHERLANDS
29
Issue protocol first published Review first published Date of most recent amendment Date of most recent SUBSTANTIVE amendment What’s New Date new studies sought but none found Date new studies found but not yet included/excluded Date new studies found and included/excluded Date authors’ conclusions section amended Contact address
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
Page 32: E-mail: m.laurant@hsv.kun.nl Tel: +31 24 361 9640 Fax: +31 24 354 0166 DOI Cochrane Library number Editorial group Editorial group code 10.1002/14651858.CD001271.pub2 CD001271 Cochrane Effective Practice and Organisation of Care Group HM-EPOC GRAPHS AND OTHER TABLES Figure 01. Trial Flow
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
30
Page 33: Analysis 01.01.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 01 Physical function (better vs not better)
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 01 Physical function (better vs not better) Study Intervention n/N Chambers 1978 Kinnersley 2000 Shum 2000 Total (95% CI) 181/296 401/484 550/672 1452 Control n/N 298/569 450/529 546/661 1759 Relative Risk (Fixed) 95% CI Weight (%) 17.2 36.3 46.5 100.0 Relative Risk (Fixed) 95% CI 1.17 [ 1.04, 1.32 ] 0.97 [ 0.92, 1.03 ] 0.99 [ 0.94, 1.04 ] 1.02 [ 0.98, 1.05 ]
Total events: 1132 (Intervention), 1294 (Control) Test for heterogeneity chi-square=8.39 df=2 p=0.02 I² =76.2% Test for overall effect z=0.79 p=0.4
0.5
0.7
1
1.5
2
Favours Control
Favours Intervention
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
31
Page 34: Analysis 01.02.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 02 Patient satisfaction
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 02 Patient satisfaction Study N 01 Sub-category Kinnersley 2000 Shum 2000 Venning 2000 Total (95% CI) 544 635 608 1787 p<0.00001 77.90 (10.72) 78.60 (16.00) 4.40 (0.46) 596 657 571 1824 74.05 (10.78) 76.40 (17.80) 4.22 (0.54) 31.4 36.1 32.5 100.0 0.36 [ 0.24, 0.48 ] 0.13 [ 0.02, 0.24 ] 0.36 [ 0.24, 0.47 ] 0.28 [ 0.21, 0.34 ] Intervention Mean(SD) N Control Mean(SD) Standardised Mean Difference (Fixed) 95% CI Weight (%) Standardised Mean Difference (Fixed) 95% CI
Test for heterogeneity chi-square=10.78 df=2 p=0.005 I² =81.5% Test for overall effect z=8.24
-0.5
-0.25
0
0.25
0.5
Favours Control
Favours Intervention
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
32
Page 35: Analysis 01.03.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 03 Scheduled return visits
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 03 Scheduled return visits Study Intervention n/N Kinnersley 2000 Shum 2000 Venning 2000 Total (95% CI) 222/652 92/790 236/634 2076 Control n/N 168/716 79/582 161/648 1946 Relative Risk (Fixed) 95% CI Weight (%) 39.0 22.2 38.8 100.0 Relative Risk (Fixed) 95% CI 1.45 [ 1.22, 1.72 ] 0.86 [ 0.65, 1.14 ] 1.50 [ 1.27, 1.77 ] 1.34 [ 1.20, 1.49 ]
Total events: 550 (Intervention), 408 (Control) Test for heterogeneity chi-square=12.23 df=2 p=0.002 I² =83.7% Test for overall effect z=5.21 p<0.00001
0.5
0.7
1
1.5
2
Control higher
Intervention higher
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
33
Page 36: Analysis 01.04.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 04 Prescription ordered
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 04 Prescription ordered Study Intervention n/N Kinnersley 2000 Shum 2000 Venning 2000 Total (95% CI) 407/652 481/736 391/641 2029 Control n/N 434/716 518/816 421/651 2183 Relative Risk (Fixed) 95% CI Weight (%) 31.3 37.1 31.6 100.0 Relative Risk (Fixed) 95% CI 1.03 [ 0.95, 1.12 ] 1.03 [ 0.96, 1.11 ] 0.94 [ 0.87, 1.03 ] 1.00 [ 0.96, 1.05 ]
Total events: 1279 (Intervention), 1373 (Control) Test for heterogeneity chi-square=2.91 df=2 p=0.23 I² =31.3% Test for overall effect z=0.10 p=0.9
0.5
0.7
1
1.5
2
Control higher
Intervention higher
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
34
Page 37: Analysis 01.05.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 05 Hospital referral
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 05 Hospital referral Study Intervention n/N Kinnersley 2000 Lattimer 1998 Venning 2000 Total (95% CI) 33/652 26/7184 11/641 8477 Control n/N 34/716 34/7308 25/651 8675 Relative Risk (Fixed) 95% CI Weight (%) 35.6 37.1 27.3 100.0 Relative Risk (Fixed) 95% CI 1.07 [ 0.67, 1.70 ] 0.78 [ 0.47, 1.29 ] 0.45 [ 0.22, 0.90 ] 0.79 [ 0.58, 1.07 ]
Total events: 70 (Intervention), 93 (Control) Test for heterogeneity chi-square=4.12 df=2 p=0.13 I² =51.5% Test for overall effect z=1.51 p=0.1
0.5
0.7
1
1.5
2
Control higher
Intervention higher
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
35
Page 38: Analysis 01.06.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 06 Attendance at Accident&Emergency
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 06 Attendance at Accident%Emergency Study Attendance at Accide n/N Lattimer 1998 Mundinger 2000 Shum 2000 Total (95% CI) 412/7184 274/800 14/675 8659 Control n/N 398/7308 172/509 13/664 8481 Relative Risk (Fixed) 95% CI Weight (%) 63.9 34.0 2.1 100.0 Relative Risk (Fixed) 95% CI 1.05 [ 0.92, 1.20 ] 1.01 [ 0.87, 1.18 ] 1.06 [ 0.50, 2.24 ] 1.04 [ 0.94, 1.15 ]
Total events: 700 (Attendance at Accide), 583 (Control) Test for heterogeneity chi-square=0.14 df=2 p=0.93 I² =0.0% Test for overall effect z=0.75 p=0.5
0.5
0.7
1
1.5
2
Control higher
Intervention higher
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
36
Page 39: Analysis 01.07.
Review:
Comparison 01 Doctor-Nurse substitution study results, Outcome 07 Hospital admission
Substitution of doctors by nurses in primary care
Comparison: 01 Doctor-Nurse substitution study results Outcome: 07 Hospital admission Study Intervention n/N Flynn 1974 Lattimer 1998 Mundinger 2000 Total (95% CI) 7/40 507/7184 68/800 8024 Control n/N 2/19 428/7308 50/509 7836 Relative Risk (Fixed) 95% CI Weight (%) 0.6 86.9 12.5 100.0 Relative Risk (Fixed) 95% CI 1.66 [ 0.38, 7.26 ] 1.21 [ 1.06, 1.36 ] 0.87 [ 0.61, 1.23 ] 1.17 [ 1.04, 1.31 ]
Total events: 582 (Intervention), 480 (Control) Test for heterogeneity chi-square=3.32 df=2 p=0.19 I² =39.7% Test for overall effect z=2.57 p=0.01
0.5
0.7
1
1.5
2
Control higher
Control higher
Substitution of doctors by nurses in primary care (Review) Copyright © 2007 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd
37
Page 40: