|
Implementing preventive services: to what extent can we change provider performance in ambulatory care. A review of the screening, immunization, and counseling literature |
Anderson L A, Janes G R, Jenkins C |
|
|
Authors' objectives To perform a meta-analysis of office-based interventions to determine their impact on three domains of preventive care: screening, immunisation, and counselling.
Searching The following databases were searched through to 1 January 1997: AIDSLINE, Cancerlit, CINAHL, ERIC, HealthSTAR, Health & Wellness Database (formerly Health Periodicals), Mantis (formerly known as CHIROLARS), MEDLINE, NAHL, NTIS and PsycINFO. Search terms included over 50 categories and phrases (e.g. preventive health services, prompting, cuing, reminders, flowsheet, prescription pads). Further details are available from the authors. Additional articles were identified by searching the reference lists of retrieved articles and contacting authors. Only English language articles were included in the review.
Study selection Study designs of evaluations included in the reviewRandomised and non-randomised controlled studies based in the USA reporting sufficient data.
Specific interventions included in the reviewOffice-based interventions aimed at increasing provider performance in an ambulatory care setting, classified into a hierarchy of four categories: feedback (defined as information given to providers about how their practice patterns or patient outcomes compare with those of peers, their prior performance or an external standard), prompting (defined as information given to providers about a desired action concurrent with a patient encounter or event, targeting a specific clinical decision point), prompting and monitoring (defined as ongoing or recurrent information concerning the delivery of a service), and combined interventions.
Participants included in the reviewProviders of preventive services including screening, immunisation and counselling, and patients requiring these services. Screening was defined as a special test (e.g. Pap and faecal occult blood tests) or a standardised examination procedure (e.g. clinical breast and rectal examinations) used to identify patients requiring special intervention. Immunisations included vaccines against childhood and adult diseases. Counselling was defined as provision of information and advice concerning behaviour that affects the risk of subsequent illness or injury (e.g. nutrition, physical activity, safe sex practices, and smoking cessation).
Outcomes assessed in the reviewStudies had to include an objective assessment of the provider's adherence to preventive services. Medical chart reviews and patients' reports of providers' adherence were included. Provider's own self-reported performance was excluded.
How were decisions on the relevance of primary studies made?Two reviewers assessed the relevancy of studies and if necessary a third reviewer independently reviewed the studies and a consensus was reached between all three reviewers.
Assessment of study quality The authors did not state that they assessed validity.
Data extraction Using a standardised form two reviewers classified the studies by preventive care area, randomisation procedures, research site, type and number of providers and patients (study population), treatment intervention, type of control or comparison group, intervention implementation, assessment period, and unit of analysis.
Methods of synthesis How were the studies combined?Studies were divided into those that used providers as the unit of analysis and those that used patients as the unit of analysis. Where patients were the unit of analysis the Pearson product-moment correlation coefficient 'r' recommended by Rosenthal was used as a measure of effect size (see Other Publications of Related Interest no.1). Where the providers were the unit of analysis and the patient level data were not available the average adherence rate per provider was calculated and effect sizes expressed as 'd' values. Weighted 'd' values were used to estimate the pooled effect sizes, with 'r' values converted to 'd' values using the Fisher z transformation.
How were differences between studies investigated?A chi-squared test was used.
Results of the review Forty-one articles reporting 43 studies with a total of 117 behavioural outcomes were included
Mean effect sizes:
a. Unit of analysis=provider.
Screening (n=23) weighted=0.14, unweighted=0.12.
Immunisation (n=4) weighted=0.18, unweighted=0.16.
Counselling (n=7) weighted=0.28 but chi-squared test showed significant heterogeneity, unweighted=0.23.
b. Unit of analysis=patient.
Screening (n=38) weighted=0.12 but chi-squared test showed significant heterogeneity, unweighted=0.20.
Immunisation (n=14) weighted=0.15 but chi-squared test showed significant heterogeneity, unweighted=0.13.
Counselling (n=9) weighted=0.08 but chi-squared test showed significant heterogeneity, unweighted=0.12.
Authors' conclusions The findings from our review suggest that although the effect sizes are modest, office-based interventions deserve continued development and evaluation for enhancing the delivery of preventive services. Our findings also point to the need for multiple interventions.
CRD commentary This is a clearly presented review based on a well-defined question. An adequate search of the literature was performed using a wide range of electronic databases and authors were contacted to locate further studies. However, only English language studies were included and so relevant data may have been omitted. A minimum of two reviewers selected studies for inclusion and extracted the study data, however the validity of the studies was not considered. The studies would appear to have been pooled appropriately considering heterogeneity between studies using the chi-squared test. However, no details were provided about the individual studies which makes it difficult to say definitively whether the synthesis was appropriate, particularly with regard to the pooling of randomised and non-randomised studies. The data presented would appear to support the authors' conclusions and implications, although caution is advised for the reasons outlined above.
Implications of the review for practice and research Practice: The authors state that 'in the absence of follow-up studies, we were unable to draw conclusions about the long-term effects on provider adherence to recommendations of preventive services'.
Research: The authors identified a lack of long-term follow-up studies of office-based interventions for increasing provider adherence to preventive services in the ambulatory care setting. The authors also identified a need for multiple office-based interventions.
Bibliographic details Anderson L A, Janes G R, Jenkins C. Implementing preventive services: to what extent can we change provider performance in ambulatory care. A review of the screening, immunization, and counseling literature. Annals of Behavioral Medicine 1998; 20(3): 161-167 Other publications of related interest 1. Rosenthal R. Meta-analytic procedures for social research. Beverly Hills (CA): Sage Publications; 1984.
Indexing Status Subject indexing assigned by NLM MeSH Ambulatory Care /standards; Counseling; Humans; Immunization; Office Visits; Practice Patterns, Physicians' /statistics & Preventive Health Services /standards /utilization; United States; numerical data AccessionNumber 11999003704 Date bibliographic record published 31/01/2001 Date abstract record published 31/01/2001 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|