As the population ages, the numbers of people developing care- and cost-intensive forms of dementia are raising. We investigated the pathways of incident dementia patients to long-term care (LTC) dependence and death, and examined the effects of: (1) the type of the main treating physician, differentiated by neurologists/psychiatrists (NPs) and non-NPs; (2) the prescription of antidementive drugs on the risk of needing LTC, differentiated by the degree of care need. Longitudinal claims data of the largest German public sickness fund of 10,043 incident dementia cases aged 60 years and above were analyzed for the years 2006 to 2010. Cox proportional hazard models were performed to investigate the risk of LTC and death based on what type of physician was treating the patient, and whether the patient was prescribed antidementive drugs; adjusted for age, gender, cardiovascular comorbidities, and the previous LTC level. The patients who were primarily treated by NPs had a significantly lower risk of LTC than patients who were mainly treated by non-NPs (considerable LTC: RR = 0.72, p = 0.000, severe LTC: RR = 0.78, p = 0.000, extreme LTC: RR = 0.67, p = 0.001). They generally had a lower risk of death. Antidementive drug treatment was correlated with an increased risk of LTC (considerable LTC: RR = 1.66, p = 0.000, severe LTC: RR = 1.50, p = 0.000, extreme LTC: RR = 1.48, p = 0.000) but with a decreased risk of death. A higher rate of involvement of specialists in the treatment of dementia patients is associated with a reduced LTC dependence and increased survival of dementia patients. Antidementive drug treatments appear to extend live years with dementia.
Keywords: Antidementive medication; cox proportional hazards models; dementia; long-term care; neurologist; psychiatrist.