Nursing consultations in rheumatology (NCR) are organizational patient-centered models of care in the field of rheumatology nursing skills. NCRs have been developed in recent years, but their implementation in Spain is still piecemeal. There are various models of NCR but there is no operational definition. Although there is little scientific evidence, one might assume that the NCR optimize each professional skill and contribute to improve the efficiency and quality of care by decreasing costs, especially medical, improving outcomes and quality of life effectiveness and perceived patient outcomes. Given the difficulty to find a definition of NCR in the existing literature and its potential relevance to quality of care, we designed this research project with the overall aim of contributing to the operational definition of an NCR and advance the exploration of those aspects of care, teaching and research of the NCR that could help measure quality.

To address this, Delphi rounds were considered the most appropriate method by experts in the field. The Delphi method was initially developed by the Rand Corporation in the United States to make provisions on military matters after World War II.It has subsequently been used in various areas, both as a predictive tool or as a prioritization strategy formulation.1,2 The objective of this technique is to improve judgments and forecasts on issues where there is uncertainty through expert opinion. It is based on the idea that the degree of consensus reached among a group of experts who are familiar with the topics under study gives a very rough idea of the actual size of these items.3 This technique is very useful to know a problem when it cannot be addressed with experimental approaches. The Delphi technique has been used in a variety of health-related studies. For example, it has been used to make predictions,4–7 prioritize design strategies,8–12 and especially to develop standards of quality of care and appropriate use of medical technology.13–20 This method was designed for this project, which aims to develop quality criteria and standards for defining and characterizing the NCR.

The project was an initiative of the Nursing Task Force of the Spanish Society of Rheumatology (GTESER), carried out by a scientific committee (CC) consisting of 10 rheumatologists and 3 nurses. In addition, two experienced researchers participated in the methodology design. The project development schedule is shown in Fig. 1.

Fig. 1.

Overview of the development of the project.

(0.24MB).

After defining the objectives of the project, we developed a specific questionnaire addressed to experts in the field, to be applied by Delphi methodology. Basically, the Delphi method is carried out through a series of rounds in each of which questionnaires are distributed with specific questions made to a panel of experts who score their opinions on a quantitative scale. In the second round, and the following if any, experts receive “feedback” on the scores of the rest of the group (in the form of the mean score for each question) and have the opportunity to review and change their score if they wish. The process ends when it is considered that it has reached a certain level of consensus or that the information obtained is sufficient. This method ensures the anonymity of responses for each respondent (each responder knows what he has responded, how many have responded to what question, but does not know who has answered what). The anonymity of responses prevents the group from being dominated by powerful or vehement individuals, which sometimes are characteristics not necessarily associated to knowledge. In addition, one of the strengths of the Delphi method is that it identifies the degree of agreement, but does not force consensus, but rather weights the views expressed by the number of respondents and the intensity of their responses.

In this project, we used the Delphi method for 2 rounds, with the distribution of questionnaires via email. To prepare the questionnaire for the first round, we constructed a list of features that could describe an NCR or standards that should be met. To do this, we started from the results and experiences of several SER projects: SERAP,21 ESPERANZA,22 EMAR23 from clinical practice guidelines of the SER24,25 and publication of standards of quality of care in rheumatology.15 The list of features or standards are grouped into the following sections: (1) initial considerations, (2) structural standards, (3) process standards, (4) treatment and monitoring, (5) health education; (6) training/research by the nursing staff of the NCR, and (7) quality of care.

This produced a first draft of the questionnaire sent to members of the CC to verify the relevance of the items, that the statement was not ambiguous and that the way to answer them was appropriate. With the contributions of the CC, the final questionnaire was designed during the first round consisting of 159 items grouped into 7 sections above, and within each section, grouped according to the type of response.

To respond to the questionnaire, a panel of 13 members of the CC and 54 experts (rheumatologists and nurses) was formed.

To rate indicators, the Delphi method was used for 2 rounds.8 Each panelist received by email, in October 2010, an Excel workbook with 3 sheets: (1) introduction and explanation of the project, (2) instructions on how to complete the questionnaire, and (3) the questionnaire of the first round. They set a deadline of 15 days in order to respond to it.

For each of the indicators, the panelists rated a number, depending on the chosen indicator. Sometimes they had to rate on a scale of 1–9 (1=not important and 9=very important), according to the importance they attached to the statement regarding this item. At other times, they were asked to quantify the value of an indicator (e.g. the maximum number of people per nurse at the NCR). To develop the questionnaire for the second round, a descriptive analysis was performed by calculating the mean and standard deviation (SD) of the responses of each indicator. This analysis produced a specific sheet in Excel format, for each of the panelists. This sheet included, for each indicator, the mean, SD and the answer the panelist gave in the first round. Therefore, there were as many different questionnaires as panelists responded to the first round. In addition, with the suggestions given by the panelists in the first round, the final questionnaire was prepared for the second round.

On December 17, 2010, we emailed the questionnaire of the second round to the panelists who responded to the first. This questionnaire consisted of 182 items. Each group included scores (mean and SD) and the actual score of the first round. Each panelist was asked to score points over each item and could maintain the same score of the first round, or modify it.13–15 The panelists were given 15 days to respond.

After receiving the scores, we performed statistical analysis of the second round of ratings. To categorize the variability in responses and its opposite (the agreement), we used the coefficient of variation (CV) categorized with the following breakpoints: CV less than or equal to 25%=very low variability (high agreement), CV greater than 25% and less than or equal to 50%=low variability (high agreement), CV greater than 50% and less than or equal to 75%=average variability (medium agreement), CV greater than 75% and less than or equal to 100%=high variability (low agreement), and greater than 100% CV=very high variability (very low agreement).

Due to the circumstances of this study, which only reflects the views of health professionals and patients, it was not submitted for review by the ethics committee.

Of the 67 panelists initially contacted for the first round of the study, responses were received from 49 (73%): 37 rheumatologists (71%) and 12 nurses (92%).

Of these, 46 panelists (34 rheumatologists and 12 nurses) from 12 Spanish autonomous communities responded to the second round (93% of respondents from the first round).

In Tables 1–7 specifies the items distributed in each domain, with their average score, SD, median, minimum, maximum, CV and degree of agreement.

In the items related to general considerations (Table 1) it is worth highlighting the very high degree of agreement in all cases. The panelists believe that an NCR is needed in the Rheumatology department and that it contributes to improving quality of care, waiting times, resource use, costs, clinical outcomes and patient satisfaction.

Regarding structural standards (Table 2), this showed less agreement on the need for a crash cart (Table 2.1, subcategory of equipment) and the maximum number of people per NCR in a health area and for every nurse (Table 2.3, subcategory of population). There was a very high agreement on standards relating to the training the nurses should have to administer different questionnaires, especially the visual analog scale (VAS) and the Health Assessment Questionnaire (HAQ) and metrology training in rheumatoid arthritis regarding the Disease Activity Score 28 (DAS28) and the Bath Ankylosing Spondylitis Ankylosing Spondylitis Disease Activity Index (BASDAI) and the Bath Ankylosing Spondylitis Functional Index (BASFI).

In the items related to process standards we found a very high degree of agreement in all those included in the subcategories of Tables 3.1–3.3, except for the minimum number of days per week that the NCR should be available (Table 3.2) and the need for an appointment reminder to the patient the previous day (Table 3.3). As can be seen in Table 3.5, regarding the time of visit standards, we found a much lesser degree of agreement according to numerous items, highlighting a very low agreement in the minimum estimated time before scheduling appointments.

The items related to standards of care and follow-up (Table 4) and health education (Table 5) had a very high degree of agreement except an agreement only measured as high in “nursing training in relaxation techniques for application to patients” (Table 5).

The items in Table 6, relating to training and research standards, had a greater variability, being medium, low or very low in those regarding days required for continuing education and auxiliary nursing, number of research projects by nursing and number of national publications, respectively. For all other items there was a very high degree of agreement.

The items related to care quality standards (Table 7) had a very high agreement, except the maximum time to resolve claims, the maximum time for clinical queries and months the clinics should distribute questionnaires of perceived quality.

To our knowledge, this is the first published study intended to clarify the features an NCR should have.

An important finding in this study was the high degree of agreement (CV<25%) achieved in most of the standards. Those related to initial considerations stand out as most important aspects considered, stating that the NCR is an essential component of Rheumatology departments and helps to improve the quality of care.

The structure standard underscores the need for the NCR to be located within the Rheumatology department. The most important minimum requirements are to have its own phone line, patient education materials, sphingomanometer, and scales to weigh and measure patients. In addition, the NCR nursing staff should be trained primarily for the performance of the VAS, HAQ, DAS28, BASDAI and BASFI.

Within the structure standards, there was also good agreement in establishing the average maximum population (about 200000 inhabitants) to justify establishing an NCR in a health area and the maximum number of rheumatologists who share an NCR nurse (3, 9). In contrast, the agreement was lower in deciding the average maximum population per nurse in an NCR (about 165000 inhabitants).

In the standard process it is worth noting the high degree of agreement that exists in almost all of the items. In the section on the NCR nursing staff, the panelists considered essential that the nursing staff managed their own agenda, kept a patient registry, applied protocols for the management of patients with rheumatologic disease and coordinated communication with the rheumatologist, physical therapist, occupational therapist, psychologist and social worker. Furthermore, the availability of NCR appointments must be at least 3 days/week.

In point of care activities of NCR, health education to patients and families, telephone assistance, global health assessment by the patient (VAS), pain VAS assessment and recording different nursing activities are considered essential. In patients with rheumatoid arthritis, functional capacity assessment by the HAQ is considered necessary. For patients with ankylosing spondylitis, the most important is the assessment of disease activity through the BASDAI and functional capacity assessment using the BASFI. For patients with osteoarthritis, the most important care activities of nursing personnel in the NCR should be to provide information on physical measures (hot-cold) and provide information to patients on an individual basis. For patients with chronic low back pain, individualized information and individualized advice on exercise and posture are the most prominent. For patients with osteoporosis this also applies, as individualized information and recommendations on adequacy of housing to prevent falls are the most necessary. For patients with lupus, individualized information stood out. In all conditions individualized information by the nurse stands over group information.

Panelists considered as important those aspects related to the registry of the NCR nursing staff. The most important item was to note either in the medical record (patient history) or in a database, patient assessment, medication sheet, activity planning, implementation of the activity and its assessment. It is of utmost importance to ensure patient confidentiality of the data. Furthermore, the patient history should always note the date and time in a clear and concise manner. The nurse's signature must appear on all medical records. All this information should also complete NCR treatment protocols.

There was also a high degree of agreement for some of the time standards, but not for the whole. Of note was the ‘very low’ degree of agreement when asked how far in advance the patient should be able to make an NCR appointment. At this point, the panelists issued responses ranging between 0.3 and 70 months. The lack of agreement on this point may be due to different mechanisms established in each department to meet the urgent demands or preference of patients, and the unscheduled tasks and availability for first appointment of different professionals (e.g. residents, physicians, nursing staff, secretary). It is desirable that these responsibilities are well defined to avoid delays or redundancies in unscheduled care.

By studying the results of treatment and monitoring standards, the majority of opinions highlighted the need for a protocol for monitoring disease modulating biological and nonbiological drugs, and applying it. It is also very important to ask the patient for their adherence to treatment at each visit.

Among the NCR health education standards, the panelists believe that it is essential to teach patients treated with subcutaneous drugs to administer them properly and train when in self-medication.

In the block of training and research standards of the NCR staff, there is little agreement among the panelists on how to quantify the value of the indicators. The agreement is very low at the time of deciding the number of days per year of training for nursing assistants, standard on which the panelists ranged from a day to 90 days (mean: 8 days). In the case of nursing training days, and showing better agreement, it scored, on average, 11 days a year. This variability may be explained by the different professionals that made up the group of experts who participated in the study, rheumatologists and nurses belonging to GTESER.

There was more consensus in deciding the research and staff training issues of the NCR. The most valuable lines of research were to obtain blood samples for clinical trials and adherence protocols. The most needed training topics were considered advances in pharmacologic and nonpharmacologic treatments, as well as metrology tools update. It is essential that, with the gradual implementation of NCR units, rheumatologists actively engage nursing staff in research tasks, which are not limited to mere procedure performance. This means minimal training in research methodology as part of the skills to be acquired by any member of the service.

Finally, there was also a low level of agreement to quantify time standards on quality of care. The agreement is low when setting the maximum time to answer any questions from a patient by the NCR. This time ranges from a minimum of one and a maximum of 72h. On the contrary, the agreement is highest in consideration to quality of care. The most important aspect of quality of care is to assess the satisfaction of patients. Among the most important aspects to consider in the satisfaction survey of patients seen in the NCR, information provided by the nursing staff, coordination with the attending physician, resolution of doubts and ease/availability for contacting the NCR stand out.

Although for this project followed the most appropriate methodology for developing such standards, the study is not without limitations. For example, a possible limitation may arise from the representation of the panelists. To avoid this limitation, we sent the questionnaire to rheumatologists and nurses across the country. In the end, panelists from 12 regions and different types of services participated, and it seems unlikely that there is any significant bias in the selection on the panelists and the results derived from their scores.

Another limitation may arise from the lack of completeness of the standards. To avoid missing potentially relevant standards, we searched all those available in previous publications, including15 those of SER itself. With the identified standards, we developed a preliminary list of those that should have an NCR. This list was corrected and extended by SER researchers who had participated in other projects such as the SERAP and ESPERANZA programs. It is therefore unlikely that this list missed any relevant standards, although it is possible that new standards may be defined with increased attention on NCR and their numbers.

Another possible limitations in the development of standards may be due to the influence of the opinions of some experts over others. In this project, influence among panelists was avoided using the Delphi method, since interactions occur in the second round to maintain the anonymity of responses. This method has proved its worth in the development of standards similar to those in this study.14,26–30

In any case, the standards developed in our study are quite consistent with recent recommendations made recently by the European League Against Rheumatism (EULAR) on the role of nurses in the management of chronic inflammatory arthritis.31

In conclusion, the definition of these standards may allow us to establish minimum desirable qualities in different dimensions, such as structure, process and outcome, both in clinical work and in research and teaching. The usefulness of these standards can be multiple, not only for health professionals but also in relationships with patients and patient organizations, institutions and health services and administration.

Project undertaken by the Spanish Foundation of Rheumatology and sponsored by Abbott.

Dr. D. Santiago Muñoz states the following conflicts of interest: advisorships with Roche, Celgene, and Pfizer. President of the Spanish Society of Rheumatology. Current Participation in clinical trials or studies sponsored by UCB, Roche, Abbott, MSD, Bioiberica, Celgene, Pfizer and the Spanish Foundation of Rheumatology. The other authors declare no conflicts of interest.