The study
Aims
The goal of the study was twofold. To gain insight into the current valence and quality of attitudes and confidence in professional competences of home nurses concerning depression, to determine whether these require improvement. To evaluate whether a minimal intervention could help home nurses detect depression more adequately in patients and their family caregivers and to refer them to general practitioners.
Design
A quasi-experimental field study was undertaken in three departments of a home nursing organization in the Antwerp region (Belgium) 2012–2013. Each department was similar in size with a monthly average of 581 (sd 70, range 502–637) patients being treated and an average of 17 patients for each home nurse (sd 3, range 14–19). Initially, briefing sessions were organized informing home nurses of their tasks. These consisted of detecting depressive symptoms in their patient population and their family caregivers (onwards we will refer to all participants as ‘patients’, unless when making explicit distinctions between patients and caregivers). To complete this task, home nurses could rely on several screening questions to confirm their suspicions of possible symptoms of depression. Two screening questions for depression were developed by Whooley et al. (1997) and an additional question was suggested by Arroll et al. (2005) to determine whether patients preferred professional support. In case these screening questions indicated that patients were experiencing symptoms of depression and that they preferred help, home nurses could offer patients the possibility to participate in an online intervention for depression or to refer them to local general practitioners (GPs).
In two departments, the briefing was preceded by a minimal intervention. Home nurses in the other department were used as a control group. Data were collected during a baseline assessment and at two follow-ups: a first one at 2 months following the initial intervention and a final one 7 months after the intervention. Furthermore, in the 2 months immediately following the intervention, process measures concerning the number of successful detections and referrals were also gathered.
Participants
All available nurses from three departments of the home nursing organization were included in September 2012. Everyone was eligible for participation, independent of their years of professional experience or educational level.
The intervention
A minimal intervention was developed in collaboration with a staff member of the home nursing organization working on the organization's care policy. In 1 hour, both the knowledge and the attitude of home nurses were targeted. The training was given by a researcher and a staff member of the home nursing organization for small groups of about ten home nurses at a time. It started with a discussion on familiar topics: their own experiences with patients with psychological problems and an overview of different psychiatric disorders and current psychiatric medications. From this point onwards, the training focused explicitly on depression, using components of a training module ‘Depressive patients in the general hospital’ by Schalenbourg et al. (2011). First points of interest were the symptoms and signs of depression, prevalence, the different subtypes and the comorbidity with physical illnesses, each of which was first explained by the trainers after which home nurses could ask questions. Subsequently, treatment possibilities were highlighted, including self-help, ehealth, physical exercise, short-term treatment, psychotherapy and psychotropic medication. These were also framed within the concept of stepped care. Finally, the threefold role of home nurses in dealing with depression was elaborated on: (1) being a confidante who is also an active listener; (2) detecting depressive symptoms and creating room for discussing these; (3) motivating for expert assistance. More specifically, video recordings of staged situations with actors showed right and wrong interpretations of these roles, providing the basis for group discussions. The training concluded with a brief recapitulation.
Data collection
Participants completed a self-report questionnaire on demographic data and their professional expertise. Furthermore, two additional questionnaires were also incorporated. Nurses' attitudes towards depression were measured using the Depression Attitude Questionnaire (DAQ, Botega et al. 1992). This questionnaire was originally aimed at GPs, but Scheerder et al. (2008) combined it with items from the Defeat Depression questionnaire (Priest et al. 1996) to make an extended and adapted version for pharmacists. In the current study, the questions are identical to this adaptation by Scheerder et al. (2008), aside from the home nurse being the protagonist instead of the pharmacist. The questionnaire consists of 23 items rated on a 5-point Likert scale ranging from 1 ‘I totally agree’–5 ‘I totally disagree’. Four subscales can be calculated using these items. The first is attitude towards treatment (eight items, serious vs. non-serious), where high scores indicate that professionals do not take depression seriously, as they, for example, tend to agree it is not a real disease and does not require professional treatment. Conversely, low scores indicate that professionals acknowledge the severity of depression and the need for professional help. The second is attitude towards the course of depression (four items, malleable vs. inevitable), where high scores indicate a more pessimistic attitude where depression is considered unlikely to be completely cured and a normal part of old age, whereas low scores indicate a more optimistic attitude including the idea that it can be fully cured. The third is attitude towards the role of home nurses (six items, accept vs. reject), where a low score indicates that home nurses acknowledge they may have an important role to play in helping patients to deal with depression, whereas a high score is not in favour of such a role. Finally, the fourth is attitude towards depressed patients (five items, positive vs. negative), where low scores have a positive attitude towards depressed patients, whereas high scores have a more negative attitude, considering depressed patients a strain on home nurses and unreliable. The internal consistency of the scales ranges from low to good (0·54 < α < 0·71), but the factor structure is nevertheless found fairly consistent across different health professionals and studies (Haddad et al. 2007). Interpretation can be done in two ways: (1) absolute, looking at the average score compared with the scale centre and the standard deviation of scores; and (2) relative, comparing to normative data. Although these are currently not available for home nurses, a comparison with other professionals, i.e. pharmacists (Scheerder et al. 2008), is still possible. For the evaluation of confidence in professional competences and perceived skills in dealing with depressive patients, we selected two items from the Morriss' Confidence Scale (MCS, Morriss et al. 1999). Following a Dutch back-translation, we selected two items scoring on a 10-point Likert scale ranging from 0 ‘Not at all confident’–10 ‘Very confident’. The first item, recognizing, focuses on how home nurses estimate the confidence in their own professional competences in recognizing (symptoms of) depression. The second item, motivating, is aimed at gaining insight into whether home nurses perceive they have sufficient capacities to be understanding of patients suffering from depression and whether they could motivate them for seeking help. To facilitate the interpretation of both questionnaires (of which the scale ranges vary substantially), all means were converted to POMP-scores. Such a conversion to the per cent of maximum possible scores guarantees an easier, more intuitive interpretation while retaining the integrity of the subscales (Cohen et al. 1999).
The number of detections and referrals to GPs was also monitored for 3 months following the intervention. Reporting sheets were anonymously completed by home nurses. For each potential detection, they were required to fill in the date, whether the person was a patient or a family caregiver and the person's gender. Using the screening questions by Whooley et al. (1997) and Arroll et al. (2005) mentioned earlier, they could gather the required information on whether or not the depressive symptoms that were perceived by the home nurses were actually present according to patients or family caregivers. This allowed them to determine whether first impressions were accurate.
Ethical considerations
All home nurses received informed consents prior to participating in the study. The study was approved by the ethics committee of the Faculty of Psychology and Educational Studies of the KU Leuven.
Data analysis
Data were analysed using Statistical Package for the Social Sciences (SPSS 16.0; IBM, Chicago, IL, USA). Descriptive statistics including means, standard deviations, percentages and frequencies were used to provide an overview of demographic characteristics and baseline levels of attitude and confidence in professional competences. To evaluate the evolution of home nurses' attitudes and confidence in professional competences, linear with restricted maximum likelihood (REML) were chosen, focusing on the interaction between the intervention and control group over all three measurements. This technique is particularly useful when working with unbalanced data, as it does not require each participant to provide answers at each time point to include them in the analyses. In case responses are missing completely at random at one time point, imputation can be used to estimate the value of missing data (Sterne et al. 2009).
Rigour
Trial versions of the questionnaire were provided to the home nursing organization to consider the validity, relev