Article Text

Download PDFPDF

Review: cognitive care and combined cognitive and emotional care interventions may influence patients' health outcomes
  1. Paul Haidet, MD, MPH
  1. Houston Veterans Affairs Medical Center Houston, Texas, USA

    Statistics from

 QUESTION: Do cognitive and emotional care interventions during patient–clinician interactions affect patients' health outcomes?

    Data sources

    Studies were identified by searching Medline, EMBASE/Excerpta Medica, the Cochrane Controlled Trials Register, CINAHL, PsycLIT, Amed, Sociofile, Social Science Citation Index, Science Citation Index, SIGLE, and Dissertation Abstracts databases; requesting studies from Evidence-Based Health (internet discussion list); and contacting experts.

    Study selection

    Randomised controlled trials were selected if they had ≥ 1 treatment that was a contextual intervention related to a patient–practitioner relationship and patients had a physical illness. Studies were excluded if they examined contextual factors relating to characteristics of the treatment and identified psychological interventions, or had a theoretical base (such as psychotherapy, counselling, or health education [including communication skills training]), or were directed at drug addicted or psychiatric patients.

    Data extraction

    Data were extracted by using guidelines from the National Health Service Centre for Reviews and Dissemination Guidance. Data extracted were type of non-treatment care (cognitive or emotional), country of study, number of patients, type of physical illness, outcomes, and study quality. The main outcomes were objective or subjective health status. Secondary outcomes included treatment expectations and quality of patient–practitioner relationships.

    Main results

    25 trials met the selection criteria. A meta-analysis was not done because of heterogeneity among results. 6 trials looked at cognitive care and diagnosis, where different diagnoses were given to patients with similar symptoms or different information on diagnostic testing was given. 2 studies found differences in outcomes: patients given a firm diagnosis and a good prognosis felt better at 2 weeks than did patients given an uncertain diagnosis (1 study); and systolic blood pressure was higher in patients with hypertension who were told to expect a higher reading in a second assessment than in those told the reading would be lower or not different (1 study). 19 studies examined cognitive care and treatment, studying the effect of different levels of treatment expectancy. 10 of the 19 studies reported that practitioners who attempted to influence patients' beliefs affected their health outcomes, but only 2 studies were of good quality. Practitioners' influence on patients was more effective when treatment expectations were positively enhanced. 9 studies reported no effect of practitioners' influence on patients' health outcomes: 3 studies were of very good quality, and 2 studies were of good quality. No studies examined the effects of emotional care alone, but 4 studies examined a combined cognitive and emotional care intervention. 3 of these studies showed that interventions were more effective than neutral consultations for decreasing pain and increasing recovery time.


    Patient–clinician interactions can alter patients' health outcomes.

 QUESTION: Do cognitive and emotional care interventions during patient–clinician interactions affect patients' health outcomes?


    The patient–clinician relationship has been the subject of a large amount of research over the previous 30 years.1 Previous systematic reviews of this literature have shown a consistent association between communicative behaviours and various favourable biomedical, psychological, and social outcomes.2,3

    The systematic review by Di Blasi et al adds an important dimension to these previous reviews by surveying the general medical literature for a wide range of randomised trials that included a relationship-based component as part of the intervention. 2 important aspects of this review deserve emphasis. First, a large amount of heterogeneity was present in the studies, both in the interventions and in the nature of outcomes measured, highlighting the complexity of the patient–clinician relationship and the enormous number of communicative and contextual variables that can potentially affect outcomes for any given patient. Second, the consistent and modest associations between relationship issues and favourable outcomes suggest that the patient–clinician relationship is an important modifying factor for any given treatment. This finding is not unexpected, because such important intermediate health outcomes as adherence to treatment regimens and trust in recommendations are directly related to the quality of the relationship between patient and clinician.4 The bottom line is that clinical treatments do not exist in vitro: they exist in a contextual environment that includes the patient–clinician relationship, and this relationship modifies therapeutic effectiveness.


    View Abstract


    • Sources of funding: National Institutes of Health and the John E Fetzer Institute.

    • For correspondence: Ms Z Di Blasi, Department of Health Sciences and Clinical Evaluation, University of York, York Y010 5DD, UK. Fax +44(0) 190 443 4517.

    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.