Quality of Primary Care by Advanced Practice Nurses
Quality of Primary Care by Advanced Practice Nurses
The guidelines set forth in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement were followed. PubMed, Medline and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) were systematically searched with assistance from an information specialist expert in systematic review methods to identify RCTs comparing outcomes of care provided to adults by APNs and physicians in the primary care setting. Initially, broad categories of search terms were selected, including: APN, patient outcomes, primary care and RCT. Specific terms within each category were then identified (for example, the category for APN included the terms 'nurse practitioner,' 'nurse clinicians,' 'advanced practice nursing,' et cetera). Terms were entered generally and expanded to include medical subject heading (MeSH) terms where available. Similar search terms were listed using database-specific commands such as ADJ or * (wildcard) to ensure inclusion of relevant articles (for example: family NPs, patient satisfaction, family health clinic, random* stud*). All possible combinations of the terms from each category were then searched to locate target studies (see Supplementary Appendix http://intqhc.oxfordjournals.org/content/suppl/2015/08/03/mzv054.DC1). Additional search criteria included presence of an abstract and English language publication. No date restriction was employed.
Inclusion criteria included RCTs comparing primary care outcomes of APNs and physicians, as well as any publications stemming from these RCTs providing longer term follow-up or economic evaluation of the included RCT samples. APNs were defined as nurses who had received additional formal education and training that expanded their scope of practice to include services traditionally considered to fall under the practice of medicine, such as diagnosis and treatment of medical conditions. Exclusion criteria included review articles, non-RCT design or no original data, studies in which APN care was not the independent variable and studies with entirely pediatric samples (due to the variation in outcomes of interest between adult and pediatric patients, which would impede data aggregation). Additionally, the APN role in the study had to be that of a PCP, with an educational background allowing them to manage their own panel of patients.
After removing duplicates, reviewers screened titles to identify articles that met inclusion and exclusion criteria. Next, abstracts and full text of identified studies were independently assessed by two reviewers. Discrepancies regarding whether a study met inclusion criteria and/or whether the role of APNs was equivalent to that of a PCP were discussed among reviewers until consensus was attained. Reference lists of included articles as well as previously-published systematic reviews were hand-searched for any additional RCTs satisfying inclusion and exclusion criteria.
The Cochrane Collaboration's tool for assessing risk of bias was used to assess study quality. The instrument contains seven criteria: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessor, incomplete outcome data, selective outcome reporting and other bias. Each criterion has detailed directions for making judgments about risk of bias and is rated as high, low or unclear. Two independent reviewers rated the quality of each study and discrepancies were discussed until consensus was reached among all authors.
For studies that included a cost outcome, the Quality of Health Economic Studies (QHES) instrument was also used, including 16 criteria scored as 'met' or 'not met'. Each criterion receives a weighted score ranging from 1 to 9 points totaling 0–100 points. Two reviewers (EL, AS) independently appraised each economic evaluation; where scores differed, agreement was achieved by consensus.
Data regarding study sample and setting, design and outcomes were extracted including sample size, patient characteristics and attrition rates, number and location of practices and providers, treatment description and duration, points of data collection, outcomes measured and statistical techniques. Outcomes assessed in multiple studies were then synthesized to provide comparison across studies.
Methods
Literature Search
The guidelines set forth in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement were followed. PubMed, Medline and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) were systematically searched with assistance from an information specialist expert in systematic review methods to identify RCTs comparing outcomes of care provided to adults by APNs and physicians in the primary care setting. Initially, broad categories of search terms were selected, including: APN, patient outcomes, primary care and RCT. Specific terms within each category were then identified (for example, the category for APN included the terms 'nurse practitioner,' 'nurse clinicians,' 'advanced practice nursing,' et cetera). Terms were entered generally and expanded to include medical subject heading (MeSH) terms where available. Similar search terms were listed using database-specific commands such as ADJ or * (wildcard) to ensure inclusion of relevant articles (for example: family NPs, patient satisfaction, family health clinic, random* stud*). All possible combinations of the terms from each category were then searched to locate target studies (see Supplementary Appendix http://intqhc.oxfordjournals.org/content/suppl/2015/08/03/mzv054.DC1). Additional search criteria included presence of an abstract and English language publication. No date restriction was employed.
Inclusion criteria included RCTs comparing primary care outcomes of APNs and physicians, as well as any publications stemming from these RCTs providing longer term follow-up or economic evaluation of the included RCT samples. APNs were defined as nurses who had received additional formal education and training that expanded their scope of practice to include services traditionally considered to fall under the practice of medicine, such as diagnosis and treatment of medical conditions. Exclusion criteria included review articles, non-RCT design or no original data, studies in which APN care was not the independent variable and studies with entirely pediatric samples (due to the variation in outcomes of interest between adult and pediatric patients, which would impede data aggregation). Additionally, the APN role in the study had to be that of a PCP, with an educational background allowing them to manage their own panel of patients.
After removing duplicates, reviewers screened titles to identify articles that met inclusion and exclusion criteria. Next, abstracts and full text of identified studies were independently assessed by two reviewers. Discrepancies regarding whether a study met inclusion criteria and/or whether the role of APNs was equivalent to that of a PCP were discussed among reviewers until consensus was attained. Reference lists of included articles as well as previously-published systematic reviews were hand-searched for any additional RCTs satisfying inclusion and exclusion criteria.
Quality Appraisal
The Cochrane Collaboration's tool for assessing risk of bias was used to assess study quality. The instrument contains seven criteria: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessor, incomplete outcome data, selective outcome reporting and other bias. Each criterion has detailed directions for making judgments about risk of bias and is rated as high, low or unclear. Two independent reviewers rated the quality of each study and discrepancies were discussed until consensus was reached among all authors.
For studies that included a cost outcome, the Quality of Health Economic Studies (QHES) instrument was also used, including 16 criteria scored as 'met' or 'not met'. Each criterion receives a weighted score ranging from 1 to 9 points totaling 0–100 points. Two reviewers (EL, AS) independently appraised each economic evaluation; where scores differed, agreement was achieved by consensus.
Data Extraction and Synthesis
Data regarding study sample and setting, design and outcomes were extracted including sample size, patient characteristics and attrition rates, number and location of practices and providers, treatment description and duration, points of data collection, outcomes measured and statistical techniques. Outcomes assessed in multiple studies were then synthesized to provide comparison across studies.
Source...