Article Text

PDF

Review: computer generated targeted and tailored interventions are modestly effective for improving patient health behaviour
  1. Beverly Greene, RN, MN
  1. Clinical Nurse Specialist, Ambulatory Care Regional 3 Hospital Corporation Fredericton, New Brunswick, Canada

    Statistics from Altmetric.com

    
 
 QUESTION: Are computer generated targeted and tailored interventions effective for improving patient health behaviour?

    Data sources

    Studies were identified by searching Medline (1966–99), HealthSTAR (1981–99), CINAHL (1982–99), Current Contents (1997–9), EMBASE/Excerpta Medica (1990–9), INSPEC (1969–99), PsycINFO (1967–99), Sociological Abstracts (1986–99), the Cochrane Library, Science Citation Index Expanded, Social Sciences Index, Computer Retrieval of Information on Scientific Projects, Dissertation Abstracts, internet, and LEXIS-NEXIS. Key individuals were contacted, bibliographies of articles and reviews were scanned, and key journals were handsearched.

    Study selection

    English language studies were included if they were randomised controlled clinical trials or quasi-experimental studies with evidence of instrument reliability and validity; had ≥1 patient interactive feedback, reminder, or educational intervention for improving a health behaviour; and had an association between 1 intervention variable and a health behaviour. Studies of personalised interventions were not included unless a targeted or tailored intervention was also in the study.

    Data extraction

    Data were extracted on intervention type, delivery device, health behaviour model used in the intervention, health behaviour, sample size, and results. Intervention types were personalised (person's name on the message), targeted (customised for a subgroup of the population), and tailored (messages based on the individual's characteristics). Intervention communication delivery devices were categorised as mobile, computerised, automated telephone, and print. Study quality was assessed using a 6 item rating scale. High quality was defined as ≥5 out of 10 points.

    Main results

    Of 46 studies meeting inclusion criteria, 37 high quality studies (14 targeted and 23 tailored studies) were included in the analysis. The table summarises significant findings.

    Computer generated targeted (TR) and tailored (TL) interventions showing significantly improved patient health behaviours*

    Conclusion

    Computer generated targeted and tailored interventions are modestly effective for improving patient health behaviour.

    Commentary

    The systematic review by Revere and Dunbar offers an interesting and succinct summary of the state of the art in computer generated outpatient health behaviour interventions. This review would be useful to nurses who are involved in developing ambulatory or community based interventions to assist patients in adopting behaviour changes.

    Previous reviews have focused on selected content areas. The wider lens on delivery devices, targeted and tailored interventions, and health behaviour models used in this review was an important advancement. As the philosophy of population health becomes more integrated with primary health care, nurses should consider marketing and educational strategies that will have an effect on broad community groupings.

    Although only the statistically significant findings were reported in this abstract, readers should be cautious when reviewing the actual paper because numerous improved outcomes are reported that were not statistically significant. This may lead readers to overestimate the effectiveness of the interventions.

    What is the bottom line? There is a growing body of research supporting the effectiveness of computer generated interventions. More research is needed to compare tailored and targeted interventions, multiple and single interventions, whether certain delivery devices and theoretical models are more appropriate for certain health behaviours, and costs.

    View Abstract

    Footnotes

    • Source of funding: not stated.

    • For correspondence: Ms D Revere, University of Washington, IAIMS Program, Box 357155, Seattle, WA 98195-7155, USA. Fax +1 206 543 3389.

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.