Modularity promises to relieve problems of complexity in service systems. However, limited evidence exists of its application in specialized hospital services. This study identifies enablers, constraints and outcomes of modularization in specialized hospital services. Design: A qualitative comparative study of a hematology unit with modular service architecture and an oncology unit with integral service architecture in a university hospital is performed to analyze the service architectures, enablers and constraints of modularization, and outcomes. Findings: A framework and five propositions combining the characteristics of specialized hospital services, enabling activities, and outcomes of modularization was developed. Modular service architecture was developed through limiting the number of treatment components, reorganizing production of standardized components into a separate service unit, and standardizing communication and scheduling in interfaces. Modularization increased service efficiency but diluted ownership of services, decreased customization, and diminished informal communication. This is explained by the specific characteristics of the services: fragmented service delivery, professional autonomy, hierarchy, information asymmetry, and requirement to treat all. Research limitations/implications: Modularization can increase efficiency in specialized hospital services. However, specific characteristics of specialized care may challenge its application and limit its outcomes. Practical implications: The study identifies enabling activities and constraints that hospital managers should take into account when developing modular service systems. Originality/Value: This is the first empirical study exploring the enablers, constraints and outcomes of modularization in specialized hospital services. The study complements literature on service modularity with reference to specialized hospital services.

Modularisation may support health care providers in classifying patients and delivering services according to patients' needs. However, as the findings are based on a single university hospital case study, more research is needed.

lasse lehtonen, heikki hiilamo, marina erhola, pentti arajärvi, jussi huttunen, aulikki kananoja, martti kekomäki, anneli pohjola, hanna tainio, paul lillrank, tanja saxell, katariina silander, pirkko vartiainen SOSIAALILÄÄKETIETEELLINEN AIKAKAUSLEHTI 2018: 55: 78-87 T e e m a -a r t i k k e l i

Background: The Finnish population offers many advantages for evaluating the impact of anti-dementia medication on mortality in Alzheimer's disease (AD) due to broad range of individual-level data collected in national health and social care registries and the fact that Finland has one of the highest mortality rates for dementia globally. Objective: The aim of this study was to investigate the association of anti-dementia medication with 2-year risk of death and all-cause mortality in patients with AD. Methods: This was a retrospective, non-interventional registry study based on individual-level data using Finnish national health and social care registries. An incident cohort of 9,204 AD patients (first AD diagnosis in 2012) was formed from a population of 316,470 individuals ≥74 years of age. The main outcome measure was overall 2-year risk of death. Statistical modelling was used to assess mortality (Kaplan-Meier) and adjusted hazard ratios (HR) (Cox proportional hazard model). Results: Early start of anti-dementia medication (treatment started ≤3 months from AD diagnosis) reduced significantly the risk of all-cause death compared to AD patients who had late medication initiation (defined as treatment started >3 months from AD diagnosis/no medication; HR, 0.51; 95% confidence interval (CI), 0.46-0.57). Dementia was the most common recorded cause of death in both groups. Conclusion:This study places importance on early diagnosis of AD and subsequent early initiation of drug treatment in decreasing 2-year risk of death.

After a care episode in a hospital, elderly patients often face delays in transitions to permanent residence. Poor care coordination burdens both the patients and the healthcare system. Whereas different models for coordinating geriatric patients’ care and discharge planning have been developed, evidence on their cost-effectiveness remains scant. In this study, we evaluated the associations of an integrated care model on health and social care costs and service utilization among geriatric patients admitted to a hospital in a Finnish city with c. 68,000 citizens. Elderly patient cohorts admitted before (N = 709) and after (N = 364) the implementation of the integrated care model were compared restrospectively. The new model consisted of changes in regional care criteria, discharge planning, coordination between inpatient facilities, and the daily work of nursing staff. Patients treated in the new model spent, on average, 7.4 days less in institutionalized care during one year, and the total annual cost of care decreased by 967€ per patient. A regionally coordinated care pathway from hospital admission to permanent residence may improve the cost-effectiveness of elderly care. Coordination and monitoring of outcomes at regional level is essential to avoid fragmentation of care and suboptimization among different care providers serving the elderly.