How nursing students’ placement preferences and perceptions of community care develop in a more ‘community-oriented’ curriculum: a longitudinal cohort study

The general objective of this study is to formulate recommendations for a baccalaureate nursing curriculum stimulating a positive interest in community care, based on knowledge on how students’ perceptions of community care and placement preferences develop while progressing through the curriculum.

The research questions are:

A longitudinal cohort survey study was carried out in a single institution. Nursing students underwent a redesigned curriculum with extensive elements of community care.

To influence students’ perceptions, a psychological model was chosen, describing how individuals respond to stimuli (here, the curriculum-redesign), namely affective and/or cognitive. Affective responses are fluctuating positive or negative emotions varying in intensity, and cognitive responses are ideas about how attractive a specific object (e.g., community care) is seen to be [30]. This perspective of affective and cognitive responses fits in well with the DAGMAR model - Defining Advertising Goals for Measured Advertising Results, a model for marketing and communication [31]. This model explains how individuals perform in decisions about purchases in four steps, which, in this example, is a process that begins with (1) the awareness about the possibility of a career in community care, moves to (2) an understanding what the career brings with it, then (3) to the belief about the career being the right choice, and (4) ends with a practical action: making the career choice. During this process, people progress through three successive phases: cognitive (thinking), affective (feeling), and conative (doing). Therefore, the approach of implementing a more community-care-oriented curriculum implies that students will have increased knowledge about this area (cognitive), possibly leading to a growing appeal, due to a sense of the field’s attractive aspects (affective), and subsequently a choice for a future career in this area (conative).

Nursing students from one University of Applied Sciences in a larger city in the Netherlands took part in the study. Longitudinal data collection took place four times from a single cohort of students in the full-time programme between October 2014 and May 2018 (the duration of the Dutch bachelor of science nursing programme is 4 years). Students in other educational pathways and/or specific programmes, and students that underwent only a part of the intervention due to enrolment in year two were excluded. The students were approached during allocated class time and, if not present, individually by email, in order to increase response rate. Data on students’ perceptions and placement preferences were collected with a questionnaire (see below).

The Ethical Review Board of the Open University of The Netherlands approved the study (reference U2014/07279/HVM). Students received information about the research project via the digital learning environment of the university. A letter was published about the purpose of the project, highlighting ethical aspects such as confidentiality of the information, and who would have access to the data. It was also emphasised that non-participation would in no way impact their studies. This was repeated during the data collection in the student groups, and verbal informed consent was given by all respondents.

For the survey, the Scale on Community Care Perceptions (SCOPE), a valid and reliable instrument (Cronbach’s α = .892), was used [32, 33]. SCOPE measures students’ perceptions of community care, placement preferences and underlying assumptions in 35 items. The first part of the instrument consists of items on potentially influential demographics. Students’ perceptions are measured in 33 items in three subscales: (1) the affective component of community care perception, (2) perception of a placement in community care, and (3) of community nursing as a profession. Items range from 1 (negative adjective) to 10 (positive adjective). The placement and profession subscales are aspects of cognitive components of students’ perceptions of community care. These two scales contain the extra option ‘I don’t know’, as it is relevant to gain an insight on aspects that students think they lack knowledge.

Finally, the student chooses a preferred area for a current placement from six healthcare areas: medical rehabilitation, mental healthcare, care for mentally disabled, community care, elderly care, and the general hospital. Then three aspects from the earlier profession scale that primarily determine the preference for the chosen area are selected. As this questionnaire measures the affective and cognitive component of perceptions [34], it is well aligned with the aforementioned DAGMAR-model.

The renewed curriculum, designed with the purpose to stimulate a positive interest in community care, contained an approach based on three elements: (1) influence of lecturers, (2) students’ experiences in the practice of the profession during placements and (3) new theoretical themes in the in-school curriculum.

With regard to the first element, the intervention focused on communication of lecturers to students about different areas. In a workshop to prepare curriculum redesign, reflecting on their own perceptions, many lecturers noticed they implicitly or explicitly advocated their own professional history (often related to hospital care) as a reference point. Therefore, new lecturers were recruited with a lot of expertise and/or experience in community nursing, acting as role models. Guest-lectures performed by community nurses were organised with challenging patient cases from the daily professional practice.

Second, management representatives from school and community-care organisations collaborated to ensure that a placement in the field was considered a positive experience for students. Mentors with a suitable level of education, i.e., a bachelor’s degree, are crucial in ensuring that students meet their learning needs [35‐37]. However, as a result of labour market shortages, good mentors are not available everywhere.

Third, a new in-school curriculum was implemented. To understand the structure of this curriculum-redesign, an image, by and large, of the 4-year curriculum, is needed. Year 1 and 2 include general/broad theory about all types of patients in different contexts. In year 2, all students choose a one semester/ 20 week minor-programme (30 EC) for year 3, based on a specific theme or healthcare area (e.g., health technology, mental health, community care). The other 20 weeks in year 3, and the first 30 weeks in year 4 contain two different placements, with the placement in year 4 in a, by the student, preferred healthcare area. As a consequence, students have the opportunity to create an individual pathway in their study in the last 2 years of their study, based on their own interests and mostly relating directly to their career preferences.

The first purpose of the new programme (see A in Fig. 1) was to broaden students views on the nursing profession, convincing them that nursing is more than hospital care. To get a sense of how/ which patient cases in the lessons were presented, all course materials were analysed. It appeared that, although many of these cases did not refer to a specific healthcare area, more than 60 of 110 cases took place in the hospital, compared to only four in community care. This imbalance, being an aspect of the ‘hidden curriculum’, was corrected by adding more patient cases in community care. By this time, a new more ‘community-oriented’ national Dutch profile for bachelor nursing education was developed [3]. Five themes from this profile were integrated in the broad theory programme of the curriculum: (1) fostering patient self-management, (2) shared decision-making, (3) collaboration with the patients’ social system, (4) healthcare technology, and (5) allocation of care.

In year 1 and 2 (period A in Fig. 1), the specific elements in the new curriculum were:

The second purpose of the curriculum redesign was to create a ‘paved way’ in the direction of a future career in community care, by offering a challenging in-depth programme in year 3 and 4 (period B in Fig. 1):

Four moments of data collection between 2014 and 2018 (T0, T1, T2 and T3, Fig. 1) were carefully planned to measure the effect of specific interventions. T0 and T3 were planned as close as possible to the beginning and the end of the educational programme. T1 was planned shortly after the community care week in year 2 (intervention D), and T2 at three-quarters of the third year when all third-year students by that time had experienced a placement and (at least a part of) the minor-programme of their choice.

Descriptive statistics were used to summarise student characteristics. The five negatively formulated items in the affective component scale of the SCOPE scale were recoded. Dimension scores were calculated by adding up all items and dividing them by the number of items, in the total scale and subscales, in order to give equal weight to each scale [38]. The normality of the data distribution was assessed, showing that assumptions for using parametric statistics were fulfilled. The student information system (SIS) was used for data on students’ factual placements. Students’ placements preferences at cohort level from T0 to T3 (percentage, n) were analysed using contingency tables. To examine whether and how students’ placement preferences for community care and the hospital fluctuate during their study, students’ preferences were analysed at the individual level and visualised in schemes (see Appendix 1 for the visualisation schemes of placement preferences with regard to the other four healthcare areas).

To analyse students’ perceptions of community care, first, descriptive statistics (mean, SD) were used at the four time points at the level of total scale, subscale and item. Second, the development of students’ perceptions, i.e., the effect of time (T0-T3) on perceptions, was analysed using linear mixed model techniques (LMMs). This technique is specifically suitable for analysing a repeatedly measured continuous dependant variable (perceptions/ total scale), when (1) the observations over time are not independent of each other and (2) missing data occur from students not participating at all four time points due to drop out [38‐40]. The data were analysed using IBM SPSS® version 25 (IBM Corporation, Armonk, NY).

In total 281 students started in the cohort. The size of the sample decreased 35% from 281 (T0) to 183 (T3) due to drop-out, study delay for longer than one semester, and a choice for another educational pathway during study. The response rate was high: in year 1 (T0), 273 of 281 students (97%) responded, in year two (T1), 188 of 206 students (91%), in year three (T2), 166 of 184 (90%), and in year four (T3), 170 of 183 (93%). The number of students of which data were collected at all four time points was 135.

Of the 273 first-year students who filled in the questionnaire at T0, 239 (87.5%) were female and 34 (12.4%) male. Mean age was 19.8 years (SD = 2.7, range 16–44). Of the 273, 225 (82.4%) had completed general secondary education, 43 (15.8%) had completed secondary vocational education, and 5 (1.8%) had completed higher education. The number of students born in the Netherlands was 252 (92,3%). Many students had, in some way, experience with community care: 44.0% had family or friends working in the field, 18.3% (had) worked themselves in some form of community care, and 35.2% had, either themselves or a close family member, received home care (note: an overlap in positive response to these three items is possible). The students’ demographics correspond approximately with a larger sample (n = 1058) in a multicentre study in the Netherlands [6] and, therefore, can be considered to be typical for Dutch nursing bachelor programmes. However, as this University of Applied Sciences is located in the multicultural environment of a large Dutch city, the number of students born in the Netherlands in the present sample is somewhat lower (92.3% versus larger sample 95.8%).

On the question: “If you were to begin your practice placement next week, where would you choose to do it?”, the majority of the students chose a general hospital in year 1 (79.8%), and this remains the most popular during the 4 years of study. In the development of the placement preferences, no significant differences between male and female students were observed. Mental healthcare is the second popular area at T0 (12.8%), decreasing to 8.8% at T3. Community care is the least popular area at T0 (2.6%), fluctuates during the four time points but increases between T0 and T3, to almost the same level as mental healthcare at T3 (8.2%). This changed the rank order of preference for community care from 6 at T0 to 3 at T3 (Table 1).

The description of students’ perceptions of community care at T0-T3 (SCOPE, total scale and subscales, range 1–10 (Table 2)) shows a slight increase of the mean in the total scale and the three subscales between T0 and T1, but an overall small decline between T0 and T3 (total scale 6.66 vs 6.21, respectively), with the lowest score and the largest decline in the placement scale (6.23 vs 5.51, resp.).

At the level of the separate items, it is noticeable that the item means, specifically in the affective component scale and the profession scale, mutually differ substantially (see Table 2 in bold). In the affective component scale, the items ‘important’ (T0–3 resp. 8.48–8.58), ‘meaningful’ (8.60–8.60) and ‘good’ (8.44–8.14) have a relatively high score, while items representing a more personal attraction, such as ‘attractive’ (5.26–4.86) and ‘fun’ (5.89–5.73) score lower. The results in the profession scale show that students expect to provide care mainly to elderly patients (8.81–8.62) while carrying a lot of responsibility (8.20–8.51) in a poor occupational work environment (5.84–4.15) with little collaboration (5.57–4.48). Students also see community nursing as a low status job (5.36–5.21) with few opportunities for advancement (5.72–5.33).

The longitudinal description at item level shows that, in the affective component scale, the mean in each item after T0 hardly fluctuates over time. In the placement scale, a decline after T0 is visible, specifically in the two items ‘contact with mentor’ and ‘time to evaluate with mentor’ (mean difference 1.89 resp. 1.23, and score at T3 4.46 resp. 4.92). In the profession scale, a slight increase from T0-T3 is found in items that seem to represent two larger areas. The first is ‘contacts’, represented in the items ‘enjoyable relationships with patients’ (7.68–7.77) and ‘contact with family/kin’ (7.70–7.90). The second has a more negative connotation as it appears to relate to ‘workload’, with the items ‘physically demanding work’ (6.85–7.20), ‘work pressure’ (6.87–7.40), ‘responsibility’ (8.20–8.51) and ‘freedom of action’ (7.16–7.84). All other items decline between T0 and T3, except ‘possible health improvement’ with a very slight increase (Table 2).

To measure the effect of time on students’ perceptions of community care (SCOPE, total scale), a linear mixed model analysis (LMMs) was carried out. The type lll F-test for the linear trend of the relation between ‘time’ and ‘perceptions of community care’ shows a significant result at the 5% level [F(3, 184.308) = 19.485, p < .001, indicating that the relationship between the two variables varies significantly. The estimated marginal means of nursing students’ perceptions of community care (total scale) are 6.651 for T0, 6.694 for T1, 6.142 for T2 and 6.200 for T3 (Table 3).

In order to make it easier to interpret the regression coefficient B in the mixed model analysis, a pairwise comparison of the estimated marginal means of students’ perceptions of community care between the time points (indicating the slopes) was conducted. These values increase slightly between T0 and T1 and T2 and T3. Between T1 and T2, a significant decline is shown (Table 4).

The following schemes (Fig. 2a and b) represent the individual development of students’ placement preferences. The numbers in the vertical lines represent how many students prefer a specific field at each time point (T0–3). The arrows show how students move between the time points, and the numbers in the arrows how many students take this pathway. The results at individual level show that of the 14 students with a placement preference for community care at T3, only one had such preference at T0 (Fig. 2a). This is identical to the 7 students with a preference for community care at T0, as only 1 remains at T3 (Fig. 2b); the other 6 make other subsequent choices (3 students) or drop out (3 students). Apparently, although the preference for the field in the cohort generally increases from 7 to 14, there is no consistent choice for the field of community care. With regard to the perception of community care, the 14 students preferring community care at T3 have a mean in perceptions (total scale) of 7.66, compared to 6.21 in all students (Table 2).

Comparing placement preferences of the 14 students preferring community care at T3 (Fig. 2a) with their factual placements at T1, 2 and 3, shows that 6 students at some point during study had a placement in community care (10 or 20 weeks). The same result is visible at other time points: of the 16 students preferring community care at T1, 2 had a placement in the field, and at T2 (n = 6) also 2 students. Therefore, no relation between a factual placement in community care and a preference for the field was found.

The development of placement preferences in students choosing the general hospital at T0 (n = 218) provides a different pattern (Fig. 3). The number of students who drop-out is relatively large, which is, although undesirable, not uncommon in higher education in the Netherlands [41]. However, of the 121 students preferring the hospital at T1, 95 students remain consistent in their choice at all time points. Many students choosing a different area once return back to their earlier choice for the hospital at the following time point. In the period of graduation (T3), 111 students have a preference for the hospital, compared to 15 students choosing a different area.

Strengths of this study are its high response rate compared to many other studies, and the holistic/ broad longitudinal approach of the curriculum-redesign. Another strength is that a longitudinal study on placement preferences with quantitative data at the individual level is uncommon. Some methodological problems limit the findings and interpretations of this study. The sample was drawn from one cohort in a single institution, and nursing curricula mutually differ, which limits the general applicability of the findings. Second, the small number of students in the ‘community care pathway’ was a limitation for some statistical procedures. Third, the repeated data collections may be considered as an intervention as well, and it is not easy to properly assess the impact on students of being ‘under scrutiny’.

Some issues with regard to students’ preferences for healthcare areas, and directly related, career choice, need further exploration. Little is known about the exact moment at which the definitive choice for an area for the future career takes place, and under what influences. These questions could be explored in a retrospective qualitative study with graduating students. Also, as the evidence on the positive effect of maturity on a preference for working outside traditional settings is limited, this topic needs further exploration.