By incorporating a model based on case management for the delivery of home care services by specifically trained nurses from the Basic Primary Care Teams (BPCTs) in Andalusia (Spain), we have been able to partially verify the hypothesis that health outcomes for patients and caregivers can be enhanced. In addition, this research has highlighted a change in the pattern of utilisation of social and healthcare services at the centres studied; current trends are characterised by greater diversification – more physiotherapy and social worker interventions, as well as telephone consultations with the case manager community nurse – together with a drop in the number of home visits and consultations at the healthcare centre.
Regarding patients' characteristics, the study highlights a greater degree of dependence and overall decline of subjects in the intervention group vs. control group at the time of enrolment in the programme. It would appear that patients suffering severe functional decline who, under the traditional model, would either be kept in hospital or would be referred from one service to another in the healthcare system, may now be referred to a professional responsible for case management at the primary level of healthcare. Also, our findings appear to indicate that the services provided by case management nurses and by community nurses are genuinely complementary. So, it appears that case management needs to be activated only in the case of patients with complex conditions requiring a broad range of services and co-ordination among healthcare professionals. The system should ensure that those services – some provided by other sectors – are readily available so that any changes in management of healthcare demand do not ultimately fall back on the flexibility of informal caregivers.
The results highlight a better recruitment rate of caregivers through the new home care model. This may suggest a trend that alters the traditional relationship between the formal and informal systems of care delivery. Here, informal care ceases to act only as a "resource" available to the formal system, and becomes acknowledged as a "client", or "co-client" of the formal healthcare delivery system . The caregiver profile identified in the study, i.e. vast majority of women, patients' daughters or spouses – demonstrates the clear gender bias in caregiver role assignment; culturally, it still seems inevitable that the women most closely related to the patient should take on this role . The fact that Healthcare Services are focusing their attention on this group as clients, enhances opportunities for collaboration, increases awareness of their status of inequality and their need to access resources that may lead to eradicating this inequity.
With regard to health outcomes, the study highlights first and foremost that recovery of patients' ability to perform essential activities of daily living (Basic Activities of Daily Living) is significantly greater among patients receiving care through the new model, especially among the sub-group of patients who were discharged from hospital. Given base-line differences, even bearing in mind their greater chances of improvement, the highest recovery rate indicated by the Barthel index seems to be directly related to interventions; the study shows that case management nurses are more frequently involved when patients receive more home care services such as rehabilitation and physiotherapy, and more support resources, such as technical aid. These findings are consistent with the results reported by Britton for patients who were discharged from hospital after suffering a stroke, and for which rehabilitation at home is reportedly more effective if combined with early discharge and when home care is offered to patients at a time when functional decline is greater and their transport needs more complex .
It is hardly surprising that no differences were found between both groups in terms of the impact of the intervention on patient survival given the profile of patients included in the study and considering that the effect of home care services on mortality tends to be appreciable only in young subjects. Hence, this should not be a goal per se of home care programmes, especially when targeting elderly, immobilised patients .
As to results in the group of caregivers, the new model appears to avoid, and even to slightly reverse the natural trend towards increased caregiver burden over time, as shown by the evolution of Zarit test scores for the control group. It is quite possible that structured support, with specific interventions which are key in the new model will be efficient in curbing or delaying further aggravation of the situation [8, 31].
The results of this study highlight the new model's interesting effect on the pattern of service use. Case management seems to rationalise the use of services by bringing providers into the home care setting, such as physiotherapists or social workers, who have a lesser role in the traditional model. At the same time, the new model curbs the use of general services, such as nurse home visits – on average three visits less in the intervention vs. control group – or consultation with physicians or nurses at healthcare centres. It is very plausible that this downward trend is also linked to the possibility for patients and caregivers to access telephone assistance from case management nurses.
Among the effects of this new pattern, the lower number of caregiver visits to the healthcare centre is particularly noteworthy. For caregivers this means that they are relieved of one of their most frequent and time consuming tasks  (i.e. making the necessary arrangements at health services facilities to cover patient's needs); they also gain greater accessibility through telephone-based assistance. This drop in the number of visits to health centres provides clear proof that the demand can largely be managed using care options other than physician/nursing consultations, providing greater degrees of accessibility and satisfaction for the population.
One of the most important questions arising in the face of this new utilisation activity is its possible impact on satisfaction, since part of the demand is provider-induced, in this case by the case management nurse. When comparing the data, satisfaction was seen to be significantly higher in the group of patients receiving care under the new model vs. the control group, in line with other studies . How can we account for better outcomes and greater satisfaction achieved with fewer visits? The key seems to lie in the diversification of providers and resources, and the availability of direct telephone assistance – proactive or reactive – which to some extent has shifted in situ assistance. The literature on case management reports on a myriad of experiences with different forms of tele-care that have achieved satisfactory results in randomised experimental studies . Confirmation of these effects should be the focus of future research since there is currently no automatic information system available to shed light on the incidence, type of calls and interventions conducted through this approach.
This study has shown no effect on frequency of A&E visits or hospital admissions, much like other case management studies conducted in different settings [35–37]. However, there is no doubt that a considerable effect on readmissions, and even on mortality, is achieved when patient-specific interventions are implemented, as seen in the case of individuals with heart disease, for instance [12, 13, 15]. Still pending for the full implementation of this model is the devising of critical, specific pathways for particular patients.
Equally, no effect on institutionalisation was identified in this study, despite the fact that the benefits of performing multi-dimensional assessment of elderly patients in terms of reducing admissions to residential homes is well documented in the literature . It is quite possible that cultural tradition in Andalusia has favoured care in the family setting rather than institutionalisation because, amongst other factors, the public offering of residential homes is scarce and precarious, in comparison with other neighbouring countries.
The size of the sample obtained after removing drop-outs has prevented us from drawing conclusions on some specific sub-groups of patients receiving the service, namely terminally ill patients and some outcomes as quality of life or follow-up to 18 months.
A complex data compilation process, not always exempt from possible bias, was required in view of the need to resort to different sources of information using ad hoc procedures, and so as to avoid upsetting the normal conditions of professional nursing practice, without exerting any influence on nursing staff. The different procedures used to obtain data from patients and caregivers in the intervention group (through case manager community nurses, at subject's homes, as part of the assessment process) vs. the control group (through interviewers from the research team, members with a nursing profile, via telephone interviews) may have led to differences in handling assessment criteria, for instance, despite the fact that training for data compilation was identical for all staff taking part in this stage of the research.
To round off this section addressing the limitations of the study, it is worth pointing out that the inclusion of certain variables in relation to nurses' professional practice would have allowed for a deeper analysis of the link between interventions and outcomes, beyond the mere existence, or not, of case management as a component in care delivery. As it proved difficult to obtain data through direct observation or from nursing staff statements without incurring in bias, this idea was abandoned. This issue is to be addressed in subsequent studies by our research team using a qualitative approach.