Medical information delivered to patients: Discrepancies concerning roles as perceived by physicians and nurses set against patient satisfaction

2.1. Context 

The Nantes University Hospital (NUH) employs 9000 healthcare workers (HCWs), and has over 3200 beds spread over seven sites, covering all areas of care, both acute and chronic. NUH was accredited in September 2002, with several recommendations. To respond to these new requirements, quality improvement programmes were created, and one of them relates to the improvement of medical information delivered to patients by HCWs. A multi-professional working group, which also included patients, was first of all set up. Prior to the implementation of improvements, an initial evaluation of roles and practices in delivering medical information as perceived by physicians was implemented alongside a study of patient opinions on the quality of the information they received. This work was completed by a third assessment, targeting roles and practices as perceived by nurses, in order to enable comparison with results for the physicians. The results obtained led to the definition of a concrete plan of action (drafting of a guide for patient information addressed to all the physicians and nurses in NUH, and improvement in the quality and quantity of written information delivered to patients). These two evaluations of perceived practice were the first large-scale experiments concerning HCWs in NUH.

2.2. Studies 

The first study, focused on physicians’ perceptions of information given to patients, was carried out in April 2003. All physicians, working full-time or part-time in NUH (n=794) received by mail an anonymous questionnaire to complete. One year later, a second study, focused on nurses’ perceptions of, and role in, patient information, was carried out in the same conditions in June 2004. The nurses (n=2411) were sent the same anonymous questionnaire as the physicians, comprising an additional item on perceived obstacles to information-sharing between physicians and nurses. All the questionnaires, dispatched by post, were accompanied by an explanatory letter and an envelope for the answer.

The patient satisfaction study was carried out in April 2003 in the NUH. All French-speaking patients aged 18 and over, able to complete a questionnaire, and hospitalized full-time for at least 24h in medical, obstetrical or surgical wards, were included consecutively (1600 questionnaires were distributed). Children, patients not able to answer and outpatients were excluded. Patients were approached the day before discharge by independent, trained, research assistants. These assistants explained the purpose of the study, invited patients to take part and gave them a questionnaire. Patients completed the anonymous questionnaire alone, and returned it to the assistants in a sealed envelope.

2.3. Questionnaires 

The questionnaire for use with professionals, designed by the members of the working group, explored perceived roles in the delivery of information to patients by physicians and nurses for the various phases of health care provision in hospital; it also explored the modes of delivery of information to the patient, and difficulties encountered in this respect. Opinions on the needs of patients in terms of information were also collected. The questions relating to perceived roles in each phase of patient care (diagnosis, investigations and treatment) were identical in the two studies. Finally an open-ended question enabled remarks to be made. Questionnaires were pre-tested by five physicians and five nurses, respectively and adjusted according to the results of this test.

The questionnaire for completion by patients was a standardised questionnaire widely used in French hospitals (EQS-H) [24]. The 26-item EQS-H comprises two subscales: “quality of medical information” (MI) (13 items) and “relationships with staff and daily routine” (RS) (13 items). In the validation study, the variance explained by the two factors was 42.3% and Cronbach's α was respectively 0.88 and 0.87 for MI and RS subscales. From the MI items, we selected eight items on patient satisfaction with information received, exploring the same dimensions as for professionals, and for each phase of care provision (diagnosis, investigations and treatment), with the exception of prognosis. Patients were asked to evaluate the quality of information delivery during their hospital stay on a 4-point scale (1=very satisfied, 2=quite satisfied, 3=not really satisfied, 4=not at all satisfied) with a “not concerned” response choice. A final open-ended question enabled patient comments to be made.

2.4. Statistical analysis 

Statistical analysis was performed with the usual techniques of descriptive statistics (frequency, mean and standard deviation), and analysis of contingency tables. Chi2 tests were used to compare percentages or paired Chi2 tests to compare theoretical and observed practice. The required degree of significance for each of these tests was 5%. Analyses were performed with the statistical package SPSS software (Statistical Package for Social Sciences; 11th version, Chicago, USA).

3.1. Study samples 

Of the 794 physicians who received a questionnaire, 302 completed and returned it (participation rate: 38.0%). There was a significant difference in participation between full-time physicians and part-time physicians (44.3% vs. 21.2%, p<0.01). The mean age of the physicians was 44.1 years (S.D.: 8.9), 57.1% of them were male.

Of the 2411 nurses who received a questionnaire, 533 completed and returned it (participation rate: 22.1%). The mean age was 39.3 years (S.D.: 9.0) and 87.8% of them were female.

The mean age and gender-ratio were not different from the general distribution of age and gender in the NUH. All the NUH sites were represented.

For the 1600 questionnaires issued to patients, there were 1246 respondents (participation rate: 77.9%). Fifteen percent (n=245) of patients were too ill to participate, 1.6% (n=26) did not speak or read French, and 5.2% (n=83) refused to answer. The mean age was 53.9 years (S.D.: 21.9) and 58.7% were female.

3.2. Patient information on diagnosis (Table 1) 

Both physicians (85%) and nurses (92%) considered that the announcement of the diagnosis was made by the physician alone. Likewise, information on prognosis was thought by all the professionals to be mainly given by the physician. In contrast, 75% of nurses thought that nursing staff played an important part in explaining the pathology, this result contrasting with that noted for the opinions of physicians (only 31% considered that nurses had an important role).

Table 1. Medical information towards inpatients: opinions and practices of professionals, and patient satisfaction

		Professionals						Patient satisfaction (n=1246) (%)
		Physicians, opinion on who gives the information (n=302) (%)			Nurses, opinion on who gives the information (n=533) (%)
		Physician alone	Nurse alone	Physician and nurse	Physician alone	Nurse alone	Physician and nurse	Very satisfied	Quite	Not really	Not at all satisfied
	Diagnosis
	Announcement	85	5	10	92	1	7	86	7	5	2
	Explanations	65	4	31	21	4	75	78	14	6	2
	Prognosis	77	4	19	66	2	32	–	–	–	–
	Investigations
	Objectives	65	3	32	41	4	55	86	8	4	2
	Sequence	44	9	47	9	26	65	77	13	6	4
	Benefits and risks	77	3	20	46	6	48	63	15	11	12
	Results	79	2	19	58	2	40	63	18	11	9
	Treatments
	Alternatives	78	1	21	72	1	27	61	14	11	14
	Benefits and risks	78	1	21	44	5	51	57	15	13	15

% rounded off to next whole number.

Patients generally considered that they had been adequately informed (more than 75% were “very satisfied”) on diagnosis, and the explanations provided on the pathology were considered satisfactory.

3.3. Patient information on investigations (Table 1) 

Roles in delivering information on medical investigations and tests were perceived differently by physicians and nurses. Whether this related to information on the aims of investigations, the way they were conducted, the benefits, the risks, or the results, nursing staff considered that they played an important part, in collaboration with the physician, while physicians considered that they were the main players in the information provided, except for information on the way in which complementary examinations were to be conducted. In addition, 26% of nurses considered that they were the only ones providing information on the way in which investigations were to be conducted, this perception being completely absent for the physicians.

Patients considered they had been adequately informed (more than 75% “very satisfied”) on the aims of investigations and the way in which they were to be conducted. However more than 20% were not satisfied with information delivered on benefits and risks of complementary examinations, nor with information on results.

3.4. Patient information on treatment (Table 1) 

Explanations on therapeutic alternatives were thought by professionals overall to be the task of the physician. However 78% of physicians considered that they were the only ones delivering information on benefits and risks of treatment, while more than half the nurses thought that they played a part in this information.

More than a quarter of the patients considered that the information that they were provided with on different possible treatments and on benefits and risks was not satisfactory.

3.5. Discrepancies between role conception and role in practice of physicians and nurses in delivering information to patients, according to report by nurses (Table 2) 

For each phase of care provision, nurses were asked to determine whether their daily practice in the field of information to the patient appeared to them to be coherent with their role conception or not. Differences between role conception and role in practice were reported by nurses to be considerable for all phases of information delivery, except for the diagnosis announcement phase, which remained strictly a matter for the physician (where only 5.2% of nurses reported their actual role in practice to be greater than their representation of that role). In contrast, nearly 34% of nurses reported that their role in practice in giving information about the way in which investigations were to be conducted was more extensive than their representation of that role, and likewise 31.3% of them considered that their role in practice in providing explanations of the diagnosis overstepped their representations. The majority of nurses considered that the information provided to the patients should be more the job of the physician than it is in practice at present.

Table 2. Comparison of role conception and role in practice of nurses, according to nurses

		No. of respondents	No. of responses where role conception and role in practice are alike	No. (%) of responses where role in practice is more frequent than in role conception	No. of responses where role in practice is less frequent than in role conception	pa
	Diagnosis
	Announcement	523	476	27 (5.2)	20	NS
	Explanations	520	338	163 (31.3)	19	<0.001
	Prognosis	503	376	97 (19.3)	30	<0.001
	Investigations
	Objectives	479	332	122 (25.5)	25	<0.001
	Sequence	494	309	166 (33.6)	19	<0.001
	Benefits and risks	475	338	107 (22.5)	30	<0.001
	Results	487	363	102 (20.9)	22	<0.001
	Treatments
	Alternatives	477	380	70 (14.7)	27	<0.001
	Benefits and risks	481	353	102 (21.2)	26	<0.001

NS: non-significant matched difference.

a Matched χ2.

3.6. Communication methods 

Physicians and nurses considered generally that information was always provided verbally, and around one third reported that verbal information was accompanied by a booklet or hand-out in different situations: to give explanations on the diagnosis (37%), to explain the sequence of investigations (31%), to explain benefits and risks of investigations (33%), to inform patients about results of investigations (30%), to explain treatment (34%) or to explain benefits and risks of treatment (35%).

3.7. Difficulties encountered by nurses in delivering information to patients (Fig. 1) 

The main difficulty encountered by nurses was connected with not knowing enough about the information already provided by the physician (for 55%); next, on equal footing, came lack of time, practical difficulties, difficulties connected with specific characteristics of certain patients in care, and potential reactions of patients to the information to be delivered.

View Large Image Download to PowerPoint

Fig. 1. Difficulties relating to delivery of information to patients according to nurses.

4.1. Discussion 

The present study highlights the existence of marked divergences in viewpoints between physicians and nurses with regard to the role of each in the delivery of medical information to patients. Nurses consider that providing patients with information is an integral part of their mission, involving both a specific function within care provision and a role of relaying information from the physician, to provide complementary information or re-explain what the physician said. Physicians, in contrast, have a much more restrictive view of information, generally limited to face-to-face interview. Yet in hospital settings, communication is not centred on individual interviews between physician and patient, and delivery of information to patients also involves other health professionals, each in his or her own area of competence. Several studies, in particular in the field of care provision for cancer patients, have underlined the major role played by nurses in re-wording and re-explaining, but also in informing and assisting patients in taking their part in medical decisions that concern them [25], [26]. Leiter [27] has shown the importance of the influence of nursing practices, and that this influence “is not limited to patient satisfaction with nursing care, [but also] affects patient satisfaction with care provided by doctors, information provided, and coordination of care”. The present study shows that while physicians do see that information is delivered within a healthcare team, they do not envisage a specific role for nurses, except for the delivery of information on the way in which investigations are to be conducted. One explanation for these discrepancies is the persistence of a “paternalistic” view of the doctor–patient relationship, which has been taught to them throughout their training, and which allocates the full responsibility for patient information to the physician, making him/her the only legitimate provider of such information. Another factor relates to obstacles to collaboration between physicians and nurses. A recent study concluded that “physicians are from Mars and nurses are from Venus”, noting that the lack of collaboration can be attributed to various reasons [28]. There are indeed many reasons, ranging from the nature of the nurse–physician relationship to differing work philosophies, responsibilities, social status or culture, gender inequality, and competence of nurses [20], [22], [28], [29]. Casey underlines that physicians and nurses train separately, keep separate patient records, report to different hierarchies, read different journals and use different jargon [30]. It is also possible that the physicians and nurses interpret the term “information” differently, physicians overseeing the overall care episode, and the nurses explaining how each separate act is going to occur. Finally, differences in the role representations of the different players are observed in this study at all stages in delivery of information to patients, except for the announcement of diagnosis. These results confirm those obtained in other studies [21], such as the study by Leinonen [12], which show the unsuitability of information delivered by surgeons according to nurse opinion. As one nurse remarked in the present study: “many surgeons are unable to broach the subject of the coming death of a patient, to prepare the family for the event… Nurses are often alone in coping with the death of a patient”. The present study also found a discrepancy between the actual behaviour of nurses and their own role conception in the field of delivery of information to patients. Similar results were found by Verschuren [16], particularly regarding communication between nurses and physicians.

While patient opinion on the information provided concerning diagnosis and the purpose of investigations is good in the present study, patients were less satisfied with information relating to benefits and risks of investigations and treatment, as has been shown in several other studies [12], [31], [32], [33]. Several studies have shown that lack of information is a major source of dissatisfaction [1], [34]. In the present study, as two patients hospitalized in a surgical ward remarked (free translations of the verbatim): “physicians gave me no information but nurses answered all my questions”; “more time is needed for patient information after surgery”. One patient hospitalized in a medical ward noted that: “the way physicians and nurses talk about things is not easy to follow”. And two women hospitalized in an obstetrical unit wrote: “I was worried about the lack of coordination in information delivered by different professionals”; “even if nurses were taking care of us, the physicians made only ‘brief rounds’: you are taken by surprise and do not have time to think about questions you would like to ask”.

First information to patients on serious side effects of treatment is rarely exhaustive for numerous reasons, relating to a wish to avoid distressing the patient, lack of time, but also to limited abilities on the part of patients to memorise and comprehend, even when the physician considers this same information to be essential [35], [36]. Secondly it is possible that physicians may have delivered accurate information, but that the patient was for some reason not in a receptive state of mind at the time this was done, or did not understand, or did not recall all the information. In a study on medication in depression, Bull [37] showed the existence of discrepancies between what physicians stated they had told patients and what the patients remembered. Finally, it is possible that physicians do not fully realise that exchange of information is extremely important to patients, and that nurses compensate for this shortfall in information delivery [8]. The present study confirms already published work by showing that patient satisfaction, measured with a validated tool, seems higher in areas where professional roles are clear-cut, in particular for providing information on diagnosis or therapeutic options.

The present results also show that a major problem is that nurses do not know what the patient has been told, that this information is not recorded in the patient's file, and that as a result it is difficult for them to adapt what they say to patients, which in turn has an impact on the quality of the information actually delivered. Several remarks made by nurses in response to the open-ended question at the end of their questionnaire confirm this lack of physician–nurse coordination (free translations of the verbatim):

“It is very damaging for patients and staff for nurses not to be aware of what information has been provided by the physician; this would improve the quality of care and would avoid discrediting nurses. It is sometimes children or families who give us very important information”.

“The healthcare team should ensure consistency of information; the involvement of numerous people who do not always deliver the same message can confuse the patient”.

“It would be a good thing if within each ward or department the responsibility for providing information were to be clearly defined or even codified”.

Baggs [15] also underlines a very clear need for nurses to be included more in clinical decision-making and in communication of treatment goals and plans to patients.

The present study has highlighted the predominance of verbal information. Written material is only a back-up, and is reported to be used as a complement by only a third. Yet the study by Johnson [38] has shown that the use of both verbal and written health information when communicating about care issues with patients on discharge appears to improve patient knowledge and patient satisfaction.

The present study presents several limitations: firstly the rate of participation by health professionals. While it is amply sufficient to obtain relevant data for the implementation of an improvement scheme, it cannot be seen as a reliable estimate of the perception that professionals overall in the NUH entertain, and results cannot therefore be generalised. The participation rate for physicians (40%) was similar in the study conducted by Thomas [20], but that for nurses was low. As this was the first time that nurses were interviewed on this large-scale, it may explain why they in particular were reluctant to respond. Finally the date at which the nurse survey was conducted (end of June) close to the holiday period could also be a factor that reduced the response rate. The reasons why some professionals did not answer the questionnaire remain to be explored.

The data obtained from these HCWs did not make specific reference to certain categories of patients, but to usual practices in the wards. This first study in the NUH was on the sensitive theme of information to the patient, and it intentionally avoided exploring HCWs’ perceptions of the actual quality of the information delivered to patients, focusing rather on an indirect assessment of its quality via patient satisfaction. Due to the lack of validated tools in the French-language literature, we needed to elaborate ad-hoc instruments to survey physicians and nurses, and they are not yet standardised. Finally, patients were interviewed independently from professionals, and no link can be made between their responses and those of the professionals actually providing their care. Indeed, establishing correspondences among surveys that have been conducted independently entails the loss of potential data, reducing the scope of statistical analyses. Future studies should integrate patient and healthcare staff studies to enable maximum data exploitation.

4.2. Conclusion 

The large number of HCWs who come into contact with patients means that joint reflection is required to avoid the risks of lack, shortfall or inconsistency in the information given to patients. The sharing of information within the healthcare team ensures the quality of information delivered to the patient, while conversely absence of cohesion and inconsistencies between physicians and nursing staff, and their different modes of expression towards the patient, often generate anxiety and result in insufficient information actually reaching the patient. It can be recommended first that the responsibilities of the various health professionals in the field of information to the patient should be clearly defined, and in particular that physicians should recognise the essential part played by nurses. The second recommendation is that it is important to ensure consistency in what is said to patients by facilitating transfer of information between professionals to guarantee good care coordination.

4.3. Practice implications 

In all situations, whether in face-to-face interviews with patients or within a care team, the will to communicate is essential [39]. One line of approach is therefore the improvement of communication between physicians and nurses. They could both benefit from a better definition of the roles of each, from training in conflict-resolving, in effective methods of asserting their own opinions and knowledge, in listening skills, and in conducting collaborative ward rounds. These methods for training teams in interpersonal skills have already been validated in the field of aviation or other industries, and they have shown their value in improving the quality and safety of care delivered to the patient [20]. In addition, the difficulties encountered by nurses in delivering information to patients mainly relate to lack of adequate information on the content of the information delivered by the physician, cited far more frequently than lack of time or material difficulties. It is in this area that efforts for improvement can be deployed, focusing on the need to develop a joint project implicating the healthcare staff overall, to achieve better coordination of care provision. One concrete aspect of the improvement of information-sharing among staff is the traceability of information given to the patient via his or her file.

Another line of approach relates to the training of physicians and other healthcare staff in communication techniques. Medical training often ignores cultural, social, and psychological aspects. Many initial or on-the-job training courses for practitioners have been developed, in particular in the English-speaking world, aiming to give increased importance to the relationship dimension in the training dispensed to students. This dimension is indeed central, and several studies on communication between patients and healthcare staff have shown the beneficial effect of communication on the quality of care [40], [41], [42]. Finally, future studies will explore the quality of patient participation in medical decisions that concern them, which is a subject that has not been widely explored in the French national context.

“I confirm all patients and professionals identifiers have been removed or disguised so the persons described are not identifiable and cannot be identified through the details of the story”.