Evidence Table
| Source | Communication Targets | Design Type | Study Design, Study Outcome Measure(s) | Study Setting & Study Population | Study Intervention | Key Finding(s) |
|---|---|---|---|---|---|---|
| Randomized Controlled Trials | ||||||
| Ellison 2004159 | Patients | Design Type 2 (RCT) (Level 2) questionnaire | Patient satisfaction (Level 3) | 1 hospital 85 patients | Standard care plus1 day telerounding; standard care substituting 1 day with robotic telerounding | Improvement in telerounding patients of examination thoroughness, quality of discussion, postoperative care coordination, availability of MD; in robotic telerounding improvement in availability of MD. |
| Fallowfield 2003153 | MD/patient | Design Type2 (RCT)— pre/postvideotape (Level 2) | At 12 months, same as 3 months (Level 3) | Oncology MDs, UK | 3 day residential communication skills training course | Same effect with use of leading questions, open-ended questions, and response to patient cues; improvement in fewer interruptions, increased summarizing; decline in expressions of empathy. |
| Jenkins 2002152 | MD/pt | Design Type 2 (RCT)—P-P videotape (Level 2) | At 3 months attitudes, empathy, responses (Level 3) | Oncology MDs, UK | 3 day residential communication skills training course | Improved attitudes and beliefs toward psychosocial issues compared to controls; increased expressions of empathy; open questions; appropriate responses to patient cues and psychosocial probing; self-reported changes in communication styles. |
| Joos 1996157 | MD/pt | Design Type 2 (RCT)—P-P questionnaire (Level 2) | Communication skills (Level 3), and compliance and utilization (Level 2) | 42 MDs and 348 patients with chronic conditions | 4.5 hours of training | Increased number of times MDs elicited patient and RN concerns, increased patient perception of amount of information received, no change in patient compliance with medications or appointments; no change in patient utilization. |
| Levinson 1993156 | MD/pt | Design Type2 (RCT)—P-P audiotape (Level 2) | Communication skills (Level 3) | 53 community-based MDs and 473 patients | A short CME program (4.5 hours) and a long CME program (2.5 days) | Short program: no effect. Long program: more open-ended questions, asked patient opinions, gave more biomedical information, patients disclosed more information, decrease in negative affect for both, patients had fewer signs of outward distress during visit. |
| Lozano 2004154 | MD/children (3–17) with asthma | Design Type2 (RCT)—cluster P- P interview and questionnaire (Level 2) | Asthma symptom days, asthma-specific functional status, frequency of oral steroid courses (Level 1) | 42 primary care practices in 3 locations | Peer leader education (PLE) and peer leader + nurse-mediated organizational change (PACI) | Peer leader: fewer symptom days per year & lower oral steroid rates. Peer leader + nurse: fewer symptom days per year & greater adherence to treatment by parent report. |
| Sullivan 2005155 | MD/children (3–17) with asthma | Design Type 2 (RCT)—cluster P- P interview and questionnaire (Level 2) | Symptom-free days (SFDs); asthma-related health care costs (Level 1) | 42 primary care practices in 3 locations | Peer leader education (PLE) and peer leader + nurse-mediated organizational change (PACI) | SFD: 6.5 with PLE vs. usual, 13.5 with PACI vs. usual; compared with usual incremental cost effectiveness ratio was $18/SFD gained for PLE and $68/SFD gained for PACI. |
| Tran 2002158 | ED patients | Design Type 2 (RCT) (Level 2) questionnaire | Patient length of stay (LOS), wait time, perception of LOS, ratings of nurse skills and MD skills (Levels 3 & 4) | 1 hospital ED 619 patients | Providing patients with information q 15 minutes during stay | No difference in LOS, wait time, nurse skills. Decrease in perceived LOS and wait time and increase in perception of MD skills. |
| Nonrandomized Controlled Trials and Quality Improvement (QI) Projects | ||||||
| Boyle 20044 | MDs/RNs | Design Type 6 P-P 2 units no control (Level 5) | Communication skills, increased staff satisfaction, lower stress, increased problem-solving using videotape vignettes, questionnaire | 1 ICU from 2 hospitals | Collaborative Communication Intervention over 8 months: 23.5 hours for 6 modules | Increased perceived RN and MD communication skills, improved nurse leadership and problem-solving, decreased staff nurse personal stress. |
| Copnell 2004134 | MDs/RNs | Design Type 6 P-P 2 units no control (Level 5) | Perception of collaboration | 2 NICUs | Added NP | No difference before and after NP; MDs and RNs disagreed about collaboration with MDs scoring higher. |
| Dechairo-Marino 2001166 | RNs | Design Type 6 action research— P-P 1 group-no control (Level 5) | RN reports of collaboration with MDs and RN Satisfaction with decisionmaking process- (Level 3) | 1 university teaching hospital; RNs in 3 med-surg units and 2 ICUs | Activities to promote interdisciplinary teamwork between MDs/RNs, including developing principles, discussion in meetings, 1 4- hour class on decisionmaking | No differences |
| Dutton 2002170 | MDs, nurses, patients discharge planners | Design Type 8-no control group (Level 5) | Patient volume, LOS, ED closure (Level 3, 4) | 1 hospital trauma service | Daily discharge multidisciplinary rounds | Increase in patient volume, decrease in LOS, decrease in ED closure. |
| Lassen 1997169 | Well-newborn nurses, pediatricians, neonatologists | Design Type 13- QI project with a control time (Level 5) | # of admissions with R/O sepsis, LOS, # of doses of antibiotics, costs, # of readmissions, reduction in practice variation (Levels 1, 2, 3) | 1 tertiary hospital | Collaborative practice decisionmaking protocol development; education | Decrease in # of R/O sepsis diagnosis, decease in % of patients treated with antibiotics, decrease in patient days, decrease in costs, decrease in readmissions. |
| Leonard 2004168 | Various groups in Kaiser Permanente | Design Type 14- QI project-no control (Level 5) | Improve communication and teamwork by standardized communication (Level 3) | different groups of MDs and RNs | Introduce standardized communication methods such as SBAR, assertion, checklists, critical event training, and briefings | Standardized briefings related to reduced wrong-site surgery, decreased nurse turnover, improved employee satisfaction, improved teamwork climate, communication, and taking responsibility for errors—but few specifics provided. |
| McFerran 2005167 | Perinatal RNs, certified registered nurse anesthesists and MDs | Design Type 13 QI project-no control (Level 5) | Long-term measures: birth event data, medical-legal data, patient satisfaction data (Levels 1 & 2); short-term measures: implementation of 2–3 interventions using human factors technique during 1 year (Level 3) | 4 Kaiser Permanente medical centers perinatal staff | 4-hour human factors education program, SBAR communication technique, revising escalation policy, identifying safe communications, debriefs after adverse events, multidisciplinary reports, assertion, just culture statement (Level 3) | No long-term measures reported; 4 sites met short-term expectations for only communication initiatives. |
| Roberts 1976163 | Hospital employees (non-MD and nonsupervisors) | Design Type 3— NRCT-P-P 2 groups with 1 being control (Level 3) | Employee perception of organizational communication, job satisfaction, and opportunities for innovative job behavior (Level 3) | 1 urban hospital; ED staff members | 2.5–3 hour training sessions weekly for 4 consecutive weeks | Increase satisfaction with work, pay, coworkers, job; increase perception of opportunities for innovation; increase desire for interaction with peers; and decrease in information overload. |
| Weiss 1985165 | MD/RN/consumer | Design Type 3— NRCT with 3 groups, with 2 being matched control groups (Level 3) | Belief regarding value of shared versus physician-dominated responsibility for health care and beliefs that powerful individuals influence consumer health status (Level 4) | Recruited in large urban area | Discussion of role relationships, and problems for 2.5 hours 1 evening/month for 20 months | Decline in belief in shared versus physician-dominated responsibility for health care and increase in belief that powerful individuals influence the consumer’s health status. |
| Vazirani 2005164 | Unit organization; RN, MD, residents, hospitalist, NP | Design Type 3— NCRT 2 groups with 1 being control (Level 3) | Collaboration, communication (Level 3) | 1 hospital; 1 control unit and 1 intervention unit | Added NP, hospitalist, daily multidisciplinary rounds | Perception by MDs of greater collaboration between physicians and nurses with largest effect with residents, between physicians and NPs, better communication between MDs; no difference in nurse perception of communication or collaboration between nurses and MDs, nurses perceived better communication with NPs than MDs. |
| Systematic Literature Reviews | ||||||
| Di Blasi 1996 (Cochrane Collaboration-Centre for Reviews and Dissemination)161 | Patients with various health problems | Design Type 11 structured review (Level 1) RCTs with and without placebo | Health outcome, symptom resolution, functional status (Level 1); health service use, medication adherence, anxiety, satisfaction (Level 3) | 26 studies with 3,811 participants: poor quality studies with small sample sizes | Various treatments or disease management, including labeling, changing patient expectations, combining treatment information with emotional support | Labeling: no effect; changing patient expectations: conflicting results—improved lung function with suggestion of drug effects but improved systolic blood pressure following any interaction; combined information with support: improved outcomes, mixed result—6 studies found decrease in pain with improved patient-practitioner interaction, style of interaction can influence physical health but with small effects. |
| Hulsman 1999 (Cochrane Collaboration-Centre for Reviews and Dissemination)160 | Graduate or postgraduate MDs | Design Type 11 structured review (Level 1); evaluation studies RCT and NRCT P-P video, discussion, role play, audio, written, self-rating | Receptive behaviors, information behavior, interpersonal and affective behavior, psychosocial problems and emotions (Level 3); compliance, health status, psychosocial status (Level 2) | 14 studies, 408 participants, 135 controls | Training, education using lecture, modeling, discussion, role play—4–96 hours over 2 days to 6 months | 10 studies report some training effect with best designed reporting fewest effects; improved self-rating of communication and recognition of psychosocial patient problems, no conclusive patient compliance effect, no effect on health status, ambiguous effect of psychosocial health. The other 4 studies report no effects. |
| Zwarenstein 2000 (The Cochrane Collaboration)100 | Chiropodists/podiatrists, dentists, dietitians, MDs, hygienists, psychologists, nurses, pharmacists, occupational therapists, and others | Design Type 11 structured review (Level 1); RCT, controlled before and after, and interrupted time series | Self-reported health status, disease incidence, cure rates, mortality, complication rates (Level 1); adherence, satisfaction, continuity of care, costs (Level 3) | 89 studies; none met the inclusion criteria | Interprofessional education (IPE) versus single-discipline education | No conclusive evidence of the effectiveness of IPE in relation to professional practice or health outcomes. |
| Zwarenstein 2000 (The Cochrane Collaboration)100 | MDs/RNs | Design Type 11 structured review (Level 1); RCT, controlled before and after, and interrupted time series | MD/RN collaboration/joint decisionmaking (Level 3), costs(Level 4); LOS, mortality (Level 1) | 2 studies with 1,102 admissions in one and 417 admissions in the other | Training, workshops, ward reorganization, team development, meetings, patient-centered care, 4 times weekly rounds, weekly case conference | 1st study: shorter LOS, reduced costs, no difference in mortality rate, increased staff satisfaction. 2nd study: no difference in LOS and no difference in mortality rates. |
| Descriptive | ||||||
| Aiken 1994171 | MDs/RNs | Design Type 4 cross-sectional (Level 5) | Medicare mortality rates (Level 1) | 39 Magnet hospitals, 139 controls | None | Magnet hospitals (higher autonomy, control, MD relationships, RN hours, skill mix) had lower Medicare mortality rates. |
| Aiken 199999 | MDs/RNs | Design Type 4 cross-sectional (Level 5) | 30-day mortality, patient satisfaction, nurse-patient ratios, control by bedside nurses; specialty physicians (Levels 1, 3) | 40 units in 20 hospitals; 1,205 patients and 820 nurses | None | Better nurse-patient ratios, lower mortality; higher nurse control, higher patient satisfaction. |
| Alt-White 1983105 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Nurse-physician collaboration (Level 3) | 46 units, 446 nurses | None | Primary nurse, critical care units, unit communication, coordination, nurse satisfaction associated with better collaboration. |
| Baggs 1997107 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Nurse/physician collaboration and satisfaction with decisionmaking, nurse retention (Level 3) | 3 ICUs in 3 hospital | None | Collaboration was associated with satisfaction for all but more strongly for nurses; nurse satisfaction with decisionmaking was not associated with retention. |
| Baggs 199995 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Mortality, ICU readmission (Level 1) | 3 ICUs in 3 hospitals | None | In the medical ICU, there was an association between nurse perception of collaboration and lower risk of patient death or ICU readmission; MD reports of collaboration were not associated with patient outcomes. |
| Estabrooks 200598 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5 | 30-day mortality | 49 hospitals | None | Greater nurse-physician relationships, more temporary positions, higher nurse education level, and richer skill mix associated with better 30-day mortality. |
| Kaissi 2003106 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Nurse-physician interpersonal interaction/teamwork (Level 3) | 2 hospitals | None | 78% of nurses rated experience with MDs as very low/low or adequate. |
| King 1994108 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Nurse-physician collaboration (Level 3) | 90 nurses, 40 physicians, 4 hospitals, and 2 hospital ships | None | MDs & RNs disagreed with MDs perceiving higher collaboration than RNs. |
| Knaus 198697 | MDs/RNs | Design Type 4, 8 cross-sectional with no comparison group (Level 5) | Actual and predicted mortality, coordination of care (Levels 1, 3) | 13 hospitals | None | Hospitals with less actual mortality than predicted had better coordination of care and communication between RNs/MDs and among MDs. |
| Rosenstein 200236 | RNs/MDs/executives | Design Type 4, 8 cross-sectional with no comparison group (Level 5) | Nurse-physician relationship (Level 3) | Network of hospitals; 1,200 responses from RNs, MDs, executives | None | MDs and RNs were significantly different; more RNs have witnessed disruptive MD behavior, more RNs say the disruptive behavior is important in nurse morale; nurses perceive less support for conflict; nurses perceive MDs as unaware of relationship. |
| Zimmerman 1993103 | MDs/RNs | Design Type 4, 8 cross-sectional with no comparison group (Level 5) | Strong medical and nursing leadership, collaboration, coordination, communication, mortality, LOS (Levels 1, 3) | 9 ICUs in 9 hospitals; 316 RNs and 202 MDs | None | No difference in risk-adjusted mortality or LOS between high-performing and low-performing ICUs. |
| Shortell 1994101 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | LOS, nurse turnover, technical quality of care, meeting family needs (Levels 3, 4) | 42 ICUs | None | Higher scores on leadership, coordination, communication, conflict management, associated with shorter LOS, higher technical quality of care, greater ability to meet family needs. |
| Thomas 200370 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | Collaboration, communication (Level 3) | 8 ICUs in 2 hospitals; 90 MDs, 230 RNs | None | Most MDs rated collaboration and communication as high or very high; most RNs rated it as low or very low. |
| Zimmerman 1991102 | MDs/RNs | Design Type 4, 8 cross-sectional no comparison group (Level 5) | ICU LOS, predicted hospital mortality (Levels 1, 3) | 40 hospitals | None | Lower mortality associated with better technological adequacy and work environment; shorter LOS associated with better communication, culture, coordination, conflict management. |
From: Chapter 32, Professional Communication
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.