Elsevier

Health Policy

Volume 124, Issue 4, April 2020, Pages 470-477
Health Policy

Additional reimbursement for outpatient physicians treating nursing home residents reduces avoidable hospital admissions: Results of a reimbursement change in Germany

https://doi.org/10.1016/j.healthpol.2020.02.009Get rights and content

Highlights

  • Potentially avoidable hospitalizations of nursing home residents are costly.

  • Additional reimbursements for outpatient physicians are an effective tool to reduce them.

  • Our analysis highlights a 5–8 percent reduction due to additional reimbursements.

Abstract

Potentially avoidable hospitalizations of nursing home residents are costly and may even be harmful for the residents concerned. This study analyzes whether the introduction, in Germany in 2016, of an additional reimbursement for outpatient care physicians treating nursing home residents has led to a reduction in hospital admissions. This analysis exploits the introduction of the additional reimbursement in a difference-in-difference approach, using recipients of professional home care as a control. The analysis is based on claims data from the largest German sickness fund, which provide complete information on health care and long-term care utilization for each insured person. Our analysis highlights a 5-percent reduction in overall hospital stays and an 8-percent reduction in ambulatory care-sensitive admissions as a result of the additional reimbursements. However, we found no effect for short-term hospital admissions or for admissions at night or at the weekend. We conclude that the overall health care utilization for nursing home patients seems to have improved due to an increased presence of physicians in nursing homes during daytime working hours. Thus, an additional reimbursement for outpatient care physicians seems to be an effective tool to reduce potentially avoidable hospital admissions in the nursing home sector. However, it does not appear to improve emergency care utilization, especially out-of-hour.

Introduction

Many industrialized countries have seen long-term care (LTC) dependency develop as a social risk, and measures to finance and organize (health) care provision for recipients have increasingly become the focus of policy-makers’ attention in recent years. In Germany, health care services and long-term care are financed differently. Since 1995, there is a separate compulsory LTC insurance that covers home services and nursing home care up to a certain limit. This means that health care services for patients in long-term care are paid by the statutory health insurance (SHI) while the generic long-term care is paid by the LTC insurance. For LTC, individuals have the choice between inpatient care in a nursing home and outpatient care at home. In Germany, 783,000 elderly inhabitants live in nursing homes, which represents 27 percent of all care recipients [1]. Nursing home residents often suffer from multimorbidity and are therefore at high risk of needing medical care. Consequently, nursing home residents often have a high hospitalization rate. However, the literature suggests that some of these hospitalizations are unnecessary and avoidable [2], [3], [4]. As these are costly and potentially harmful for nursing home residents [5], this raises the question of how they can be explained and thus avoided. One potential reason for this inefficiency is that different health care sectors participate in the provision of health care for nursing home residents. For example, in Germany, the nursing home sector, the hospital sector, and the outpatient sector are involved in organizing health care for nursing home residents. Independent reimbursement schemes for these sectors might create (financial) incentives to shift workload from one sector to another.

Relating to this, one important substitution relationship that might influence the number of hospital admissions of nursing home residents is that outpatient care might be substituted by hospital care. The empirical health economics literature provides some evidence for this relationship. Kopetsch and Schmitz [6] highlight a general substitutional relationship between sectors regarding hospital stays in Germany. In the context of emergency admissions, Thompson et al. [7] show that a reduction in out-of-hours access to general practitioners in the UK leads to an increase in non-traumatic hospital admissions in the hospital emergency department. Krämer and Schreyögg [8] illustrate that hospital admissions are acting as substitutes for primary care emergency services in Germany with the strongest effects for non-urgent, short-stay hospital admissions and for elderly patients.

In addition to substitution effects for emergency admissions, the literature defines one group of hospital admissions as ambulatory care-sensitive. These are conditions that are expected to be manageable in the outpatient care sector but still often lead to hospital admissions (e.g., hypertension) [9]. Several studies have illustrated that better access to outpatient physicians and a higher physician density is associated with a reduction in ambulatory care-sensitive admissions [10], [11], [12]. In addition, Sundmacher and Kopetsch [13] show that, in Germany, an increase in outpatient care treatment is associated with a reduction in ambulatory care-sensitive admissions. Very little is known about the relationship between different reimbursement schemes and ambulatory care-sensitive admissions, however. There is a large body of literature that analyses the extent to which reimbursements affect physicians’ behavior in general [14], [15], [16], [17], [18]. Experimental and empirical work suggests that physicians take own financial considerations into account, especially in deciding about the quantity of treatments. Typically they find that more services are provided under fee for service than under capitation payment [19], [20], [21], [22], [23]. Regarding ambulatory care-sensitive admissions, to the best of our knowledge, only Bindman et al. [24], illustrate that managed care in Medicaid can reduce ambulatory care-sensitive admissions. In the context of nursing homes, only Kuo et al. [25] find that Medicare recipients whose physicians treat a high share of nursing home residents are less likely to be hospitalized with ambulatory care-sensitive conditions.

Our study contributes to the existing literature in the following way: We investigate whether the introduction of financial incentives for outpatient physicians to provide more outpatient treatment in nursing homes causally reduces the number of hospital admissions of nursing home residents. This enables us to provide further insights into mechanisms to reduce hospital admissions from nursing homes, especially ambulatory care-sensitive admissions.

In Germany, before July 1, 2016, outpatient care physicians had no financial incentive to treat nursing home patients as they received no additional reimbursement. In fact, they probably had to invest substantial effort into treating them, e.g., due to home visits. On July 1, 2016, an additional reimbursement for outpatient care physicians when treating nursing home patients was introduced on top of their regular budgets. The objective of this study is to analyze whether this additional reimbursement has led to a reduction in hospital admissions. This analysis exploits the introduction of the additional reimbursement in a difference-in-difference approach, using rich claims data. The data are from the Techniker Krankenkasse, Germany's largest sickness fund, which currently insures about 10 million people. The data comprises complete information on health care and LTC utilization for each insured person between 2014 and 2017. The treatment group comprises all nursing home residents in our data. The control group consists of recipients of professional home care.

The remainder of this paper is organized as follows: Section 2 introduces the institutional background in Germany. Sections 3 and 4 outline the data and the empirical strategy. The corresponding results and the findings of the robustness checks that are performed are described in Section 5. Section 6 discusses the results. Section 7 concludes.

Section snippets

Health care provision in German nursing homes

LTC in Germany includes all persons who are permanently in need of care due to physical or mental disability. LTC insurance is part of the social insurance system. About 90 percent of the population is covered by the LTC insurance. The insurance is always offered by the same company where a resident has SHI. In contrast to the SHI, the LTC insurance does not cover medical treatment but care treatment. It provides universal coverage based on the level of disability. By 2017, each person in need

Data and descriptive statistics

Our analysis is based on rich individual-level claims data for the years 2014 to 2017. The data is provided by the Techniker Krankenkasse, the largest German sickness fund, which currently insures about 10 million individuals. Since the statutory LTC insurance is tied to the corresponding sickness fund, the data also comprises all LTC related information of individuals who are insured with Techniker Krankenkasse. More specifically, the data comprises personal characteristics (e.g., age and

Empirical strategy

To analyze the effects of the introduction of the additional reimbursement for physicians treating nursing home residents, we conduct a difference-in-difference (DiD) analysis.

Our DiD model takes the following form:Outcomemst=β·changemt+treatm+δt+γs+Xstλ+ϵmstwhich we run separately for our outcome variables of interest. Outcomemst represents these outcomes and, in the main specification, contains the logarithm of hospital admissions of group m (nursing home residents or recipients of

Main results

Table 2 displays the results of our DiD regression model of Eq. (1) in different specifications. Column 1 shows the results for a basic DiD approach by running the regression model without controlling for additional factors, including only a dummy variable for the post-period. Columns 2 and 3 display the results when we gradually include time fixed effects, state fixed effects, and the additional time-varying control variables. The last specification in Column 3 is our preferred specification

Discussion

The results described in Section 5 are in line with the stated hypothesis: when the additional reimbursements are introduced for physicians treating nursing home patients, outpatient physicians are incentivized to increase the overall accessibility of outpatient care for nursing home patients. This reduces hospital admissions. We find a strong and robust reduction in overall hospital stays of about 5 percent as a result of the introduction of the additional reimbursement. This effect goes hand

Conclusion

This study analyzes whether the introduction of an additional reimbursement for outpatient physicians treating nursing home residents reduces the number of hospital admissions. Our results suggest that the overall health care utilization for nursing home patients seems to have improved due to better accessibility to outpatient care. As a result of the additional reimbursements, physicians have increased their presence in nursing homes during working hours. This might lead to a better handling

Declaration of interest

The authors declare that they have no conflict of interest.

References (33)

  • D.C. Grabowski et al.

    Predictors of nursing home hospitalization

    Medical Care Research and Review

    (2007)
  • T. Kopetsch et al.

    Regional variation in the utilisation of ambulatory services in Germany

    Health Economics

    (2013)
  • C. Thompson et al.

    How have changes to out-of-hours primary care services since 2004 affected emergency department attendances at a UK district general hospital? A longitudinal study

    Emergency Medicine Journal

    (2009)
  • J. Krämer et al.

    Substituting emergency services: primary care vs. hospital care

    Health Policy

    (2019)
  • A.B. Bindman

    Preventable hospitalizations and access to health care

    JAMA: The Journal of the American Medical Association

    (1995)
  • Z. Ansari et al.

    Access to health care and hospitalization for ambulatory care sensitive conditions

    Medical Care Research and Review

    (2006)
  • View full text