The small scale clinical psychiatric case registers
Version of Record online: 25 APR 2014
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Acta Psychiatrica Scandinavica
Volume 130, Issue 2, pages 80–82, August 2014
How to Cite
Amaddeo, F. (2014), The small scale clinical psychiatric case registers. Acta Psychiatrica Scandinavica, 130: 80–82. doi: 10.1111/acps.12280
- Issue online: 8 JUL 2014
- Version of Record online: 25 APR 2014
Small Scale Psychiatric Case Registers (SS-PCRs) should be considered as a particular type (research oriented) of Mental Health Information Systems (MHIS); in 2005, WHO  defined a MHIS as ‘… a system for collecting, processing, analysing, disseminating and using information about a mental health service and mental health needs of the population it serves. The MHIS aims to improve the effectiveness and efficiency of mental health services and ensures more equitable delivery by enabling managers and service providers to make more informed decisions for improving the quality of care. In short, a MHIS is a system for actions: it exists not simply for gathering data, but also for enabling decision-making in all aspects of mental health system.' In the latest years, the dramatic development of information and communication technology facilitates widespread sharing and linkage of electronic health care data, including PCR. In this new scenario, PCRs have taken the role to guide and define the rules for the implementation of high-quality MHISs.
Small Scale Psychiatric Case Registers have advantages and disadvantages. We can easily recognise that, in a small scale registration system, quality of data can be regularly checked and the number of variables collected can be higher than in a large database. For example, they can include information on the clinical condition of the patients, on psychopharmacological treatments and on duration of contacts. The combination of these two characteristics, quality and quantity of data, makes small scale psychiatric case registers of great interest for researchers and policy makers.
I will try to give some example to support my point of view. The SS-PCRs are mainly used in four areas of the epidemiological research: studies on the patterns of mental health services' utilisation (included the studies of health economics); studies on the effects that sociodemographic and clinical variables have on the utilisation and the planning of services; drug-epidemiology studies  and, finally, studies that integrate geographic information with registry data using the approach of health geography . From an epidemiological perspective, when we try to study rare events such as suicide rates or the incidence of schizophrenia, SS-PCRs have low power and many limitations that are very well explained by Munk-Jorgensen and colleagues on this issue of Acta Psychiatrica Scandinavica . On the contrary, SS-PCRs show their strength when specific phenomena that require a deeper knowledge of patients’ and areas’ characteristics are studied. This is, for example, the case of patterns of care analyses. Few studies have investigated so far factors which predict inappropriate terminations (drop-out) of clinical contact with mental health services. Utilisation of the Verona Psychiatric Case Register, that is a SS-PCR since 2000 covering a catchment area of about 450 000 inhabitants (started in 1982 as a very small register covering the South Verona area of about 75 000 inhabitants), allowed to identify patient and treatment characteristics associated with dropping out of contact with community-based psychiatric services (CPS) in that area ; and to complete the data routinely collected with the register, patients were interviewed with clinical (GAF) and satisfaction (VSSS) scales. In the year after the index contact, 26.8% terminated contact with the CPS; of these, 62.9% were rated as having inappropriate terminations (the ‘drop-out’ group) and 37.1% had appropriate terminations of contact. Drop-outs were younger, less likely to be married and their previous length of contact with services was shorter; no drop-outs had a diagnosis of schizophrenia. Multivariate statistical analysis revealed predictors of dropping out.
There are other examples of the utilisation of data from a SS-PCR to detect the effects of treatment and clinical variables on patterns of care. Drukker and Colleagues  matched the PCR data with data about the level of psychosocial functioning assessed using the Health of the Nations Outcome Scales (HoNOS) and with routine outcome measures. They showed that those patients who receive flexible assertive community treatment (FACT) use more out-patient care and have better psychosocial functioning.
In SS-PCRs, it is also possible to routinely collect detailed data, for example, on duration, type and professionals involved in each single contact with patients. Using these data, cost analyses can be performed calculating costs in a bottom-up approach that is the only way to estimate costs of each patient and then to link differences between patients’ costs, or within the same patient in different phases of illness, to individual variables present in the database. Using this bottom-up approach was possible to explain 66% of the variation in costs of psychiatric care and 13% of variation in non-psychiatric medical costs. The model obtained with a SS-PCR explains a higher degree of cost variance than previously published studies and allowed in a further study to design a new, well-balanced mental health funding system, through the creation of (i) a list of psychiatric interventions provided by Italian Community-based Psychiatric Services (CPSs) and associated costs; (ii) a new prospective funding system for patients with a high use of resources, based on packages of care . A recent review , which examined the prediction of service utilisation and costs in psychiatry, confirmed that no single variable alone is able to predict costs and that clinical factors, as diagnosis, alongside other individual's personal characteristics, as gender and age, and previous use of psychiatric services, as the most consistent predictor of higher psychiatric cost, could explain the variations in costs between patients. However, most studies have explained only 25–50% of the total variations in costs, and the authors concluded that it seems plausible that the inclusion of ecological measures in predictive models, such as socio-economic status, the geographical characteristics and social cohesion of areas where patients live, could improve the explanation of the variation in psychiatric costs.
Another field that can be investigated using small scale registers is accessibility to services. Equal access to health care is a guiding principle in countries where health care is mainly provided within public health care systems; an important objective of such systems is to meet the population needs as close as possible to where people live. Access to healthcare services varies according to both non-spatial and spatial factors. Non-spatial factors encompass economic, cultural, and social issues, as well as factors related to health care organisation and network that can be studied in details only on a small area , while spatial factors concern the environmental context, the availability of facilities, the public transport and the road network structure. Their respective importance depends on the type of health care framework (e.g. general practitioners, general hospitals, etc.) and on the type of health problem considered. However, the association between geographical distance and service utilisation is complex, being an interaction between geographical proximity to services, socio-economic conditions in local communities, service provision and pathways of care [10, 11]. For dealing with such complex issues, it appears crucial to implement reasonably simple methods to assess access, when both spatial and non-spatial factors are involved . Very few epidemiological studies have been carried out until now using innovative approaches, and all of them have recognised methodological difficulties and biases that effect their results. These studies have highlighted that, in areas where care is provided by community-based systems, social and economic characteristics of the place of residence are associated with psychiatric service use by patients. However, socio-economic status and distance from hospital or other community-based services should not prevent the access to the care, but for some types of care (e.g. day care) there is a relation between service utilisation and proximity between place of residence and services .
Another important issue for psychiatric epidemiology is to find strong and widely agreed indicators of the quality of care provided by mental health services. One of these, even if open to criticism and not exhaustive, is the excess of mortality, and of avoidable mortality, among users of mental health services. SS-PCRs do not have a specific indication for such a study that have an higher power when performed with large scale databases. Anyway, when case registers last for long time, they allow, also if small, to study mortality. In the mortality studies conducted in South-Verona, the observed number of deaths from avoidable causes among users of community-based mental health services was shown to be four times greater than expected. The standardised mortality ratio (SMR) was higher for deaths preventable with adequate health promotion policies than for those preventable with appropriate health care. Males, alcohol/drug addicted and younger patients had the highest mortality. From these results, it seems important the implementation, by specialist psychiatric services, of health promotion and preventive programmes specifically targeted to psychiatric patients [14, 15]. Long lasting follow-ups, that is, 25-years in our case , allowed also to study different patterns of site-specific cancer mortality among psychiatric patients, showing an increased SMRs for cancer of the oral cavity (22.93), lymphoma, leukaemias, Hodgkin's lymphoma (8.01) and central nervous system (CNS) and cranial nerve tumours (4.75). The SMR decreased for stomach tumours (0.49). Patients with alcoholism (5.90 for larynx), affective disorders (20.00 for lymphomas) and personality disorders (28.00 for SNC) were found to be exposed to a high risk of cancer death in specific sites.
Although the diffusion of routine data collection for administrative purposes is rapidly increasing in the latest few years, and even though these data offer a great opportunity also for epidemiological research, the role of SS-PCRs is still crucial. They must represent a laboratory where specific research questions are investigated in more details and more closely to the clinical and social environment. Nevertheless, to reach this goal, some developments are requested; small scale databases should include specific information such as scores of clinical scales to assess severity of illness and needs for health and social care, pharmacological treatment data, narrative free text that could be used for qualitative–quantitative analyses using the improved power of information technologies and should also integrate geographical information with register data using an health geography approach. They must also improve confidentiality, privacy and security to reduce the risk of unauthorized access.
Finally, small scale psychiatric case registers should not be seen in isolation, but rather as elements of a local, regional, national or wider network of information sources. Consequently, PCR should be developed with due regard of the characteristics of existing systems within the health sector and sectors other than health, such as the definitions and classification systems used . The importance of a common language should be strengthened and specific glossary should be made available.
- 1WHO (2005). Mental Health Information Systems. Geneva: World Health Organization.
- 3Information systems for mental health. Epidemiol Psychiatr Sci 2009;18:1–4., .
- 6Flexible assertive community treatment, severity of symptoms and psychiatric health service use, a real life observational study. Clin Pract Epidemiol Ment Health 2013;28:202–209., , , .
- 7A predictive model to allocate frequent service users of community-based Mental Health Services to different packages of care. Epidemiol Psychiatr Sci 2010;19:168–177., , et al.
- 11Socio-economic status and geographies of psychiatric inpatient service use; places, provision, power and wellbeing. Epidemiol Psychiatr Sci 2007;16:10–15..
- 17Statistics and information systems. In: Saxena S, Esparza P, Regier DA, Saraceno B, Sartorius N (eds) Public Health Aspects of Diagnosis and Classification of Mental Health and Behavioural Disorders. Refining the Research Agenda for DSM-V and ICD-11. Arlinghton, US: American Psychiatric Publishing and Geneva, CH: World Health Organisation 2012: 165–200., , et al.