Does informal care impact utilization of healthcare services? Evidence from a longitudinal study of stroke patients

Highlights

• Informal caregivers facilitate access to healthcare services by stroke patients.

• Amount of informal care is not significantly associated with higher health costs.

• Presence of informal caregiver raises the likelihood of access to rehabilitation.

• The role of informal care should be considered in designing policies for stroke.

Abstract

Understanding the interplay between informal care and formal healthcare is important because it sheds light on the financial implications of such interactions and may result in different policies. On the basis of a major database on 532 Italian stroke patients enrolled in the period 2007–2008, we investigate whether the presence of a potential caregiver and the amount of informal care provided influences the use and the costs of healthcare services, and in particular rehabilitation, in the post-acute phase. Primary caregivers of stroke patients were interviewed at 3, 6 and 12 months after the acute event and use of healthcare and informal care were documented. The panel dataset included socio-demographic, clinical and economic data on patients and caregivers. A longitudinal log-linear model was applied to test the impact of informal care on total healthcare costs in the observation period. A double hurdle model was used to investigate the impact of informal care on rehabilitation costs. A total of 476 of stroke survivors in 44 hospitals were enrolled in the study and presence of informal caregiver was reported in approximately 50% of the sample (range 48.2–52.5% across the three periods). Healthcare costs at 12 months after the acute event are €5825 per patient, with rehabilitation costs amounting to €3985 (68.4%). Healthcare costs are significantly different between the patients with and without caregiver in all three periods. The presence of the caregiver is associated with 54.7% increase in direct healthcare costs (p < 0.01). Instead, the amount of informal care provided does not influence significantly direct healthcare costs. The presence of caregiver significantly increases the probability of access to rehabilitation services (β = 0.648, p = 0.039) while, once the decision on access is made, it doesn't influence the amount of services used. Our results suggest that informal caregivers facilitate or even promote the access to healthcare services.

Introduction

Demand for care is continuously changing both in quantity of care offered to satisfy the needs of population and in the types of care provided. The driving forces of these changes stem from trends in demographic trends and technological innovation in healthcare. Population ageing puts pressure on a society's ability to support its elderly citizens and impacts the epidemiology of diseases (Carmichael et al., 2010). New diagnostic and therapeutic strategies bring significant improvements in survival rate in many disease areas, resulting in growth of cohorts of patients with disability and thus in need of care. Changes in one form of care inevitably affect related forms of care, independently of the structure and organization of the healthcare system (Van Houtven and Norton, 2008).

Informal care, defined as assistance supplied by close relatives or friends, plays a substantial role in the total care provided to elderly and patients with disabling diseases. Given this premise, it is not surprising to witness increasing attention towards this form of care in both the scientific literature and in the policy debate. This growing attention provides valuable knowledge about the amount of informal care and the tasks that caregivers perform in various disease areas. The bulk of the health economics literature focuses on methods for estimating informal care costs (Dumont et al., 2010, van den Berg et al., 2004, van den Berg and Spauwen, 2006). A plethora of empirical studies showed that informal care is a major component of societal costs in many chronic conditions such as stroke, multiple sclerosis, mental diseases (Anderson, 1988, Andersson et al., 2003, Clark et al., 2001, Hickenbottom et al., 2002, Kerr and Smith, 2001, Lanska, 2002, Nordberg et al., 2005, Wimo et al., 2002). Contributions from other disciplines investigated physical and psychological health, and social wellbeing of caregivers (Jaracz et al., 2012, Simon et al., 2009, van Campen et al., 2012).

One area of research that has gained significant attention in the recent years is the interplay between informal caregiving and other types of care supplied by institutionalized health and social-care systems, frequently labelled as “formal care” (Bolin et al., 2008, Bonsang, 2009, Gannon and Davin, 2010, Jimenez-Martin and Prieto, 2011, Lo Sasso and Johnson, 2002, McMaughan Moudouni et al., 2012, Spillman and Pezzin, 2000, Van Houtven and Norton, 2004, Yoo et al., 2004). Empirical studies have provided mixed evidence regarding the nature and the sign of the link between the two types of care. In other words, the empirical relationship between informal and formal care has not definitely established whether the two types of care are substitutes or complements. For example, a negative relationship may result when informal care replaces (i.e., substitutes) some of the elements of formal care or make adverse future health outcomes less likely. A different hypothesis is that informal care comprises advice, encouragement and assistance to access formal care, thus giving rise to a positive relationship between the two.

Although available studies shed light on the variety of possible interactions between the two types of care in a variety of contexts, the current literature presents several important limitations.

First, the majority of studies to date focused on the relation between informal care and formal long-term services such as nursing homes or other forms of attention provided by public or private institutions (Bonsang, 2009, Gannon and Davin, 2010, Hanaoka and Norton, 2008, Jimenez-Martin and Prieto, 2011, McMaughan Moudouni et al., 2012). The relationship between the presence and amount of informal care and healthcare utilisation and costs remains largely under investigated. In fact, we are aware of only four studies that specifically investigated whether and how informal caregiving affects the use of healthcare services (Bolin et al., 2008, Condelius et al., 2010, Van Houtven and Norton, 2004, Van Houtven and Norton, 2008, Weaver and Weaver, 2014).

In a pioneer paper on the issue, Van Houtven and Norton investigated the link between informal care provided by adult children and different forms of formal care for their elderly parents and found that the effects of actual child givers vary by the types of care (Van Houtven and Norton, 2004). The results confirm the initial hypothesis of a substitution effect, i.e., that informal care (measured as the number of hours that all children assisted a parent in the past month) reduces total healthcare service utilization of the elderly, primarily by reducing the likelihood of using home healthcare and nursing home services. The only case in which informal care is a complement is outpatient surgery: adults with informal care had higher probability of receiving this type of service compared to those without (30% for parents with median amount of informal care vs. 14% for parents without). In their more recent study, Van Houtven and Norton extended their model to investigate the impact of informal care on Medicare expenditure in the US (Van Houtven and Norton, 2008). They found that informal care by children reduces Medicare expenditure for long-term and inpatient care. Bolin and colleagues used the same model developed by Van Houtven and Norton to investigate the relationship between informal care and five types of formal care in Europe (Bolin et al., 2008). Informal care resulted as a substitute to formal home care, and as a complement to doctor and hospital visits, in contrast to the estimates obtained in the US. In addition, this study aimed to assess whether the relationship between informal and formal care differs across Europe. The results obtained supported authors' hypothesis on the existence of a north-south gradient: in the regions with “strong” family ties (i.e., southern Europe) informal care is to a greater extent a substitute for formal home care.

In a study conducted by Condelius et al. (2010) informal care proved to be significantly and positively associated with the utilization of healthcare, but only in contacts with physicians in outpatient settings and not with hospital admissions. Finally, the effect of having informal support available at home on inpatient care use was recently investigated in Switzerland (Weaver and Weaver, 2014). Results suggest that availability of informal care has no impact on the likelihood of hospitalisation but does significantly reduce length of stay.

An additional gap identified in the literature refers to methodological challenges in investigating the relation between informal and formal care. It has been argued, in fact, that the lack of consistent findings on the relation between formal and informal care is explained by the complexity of its dynamics. The two types of care can occur simultaneously or can precede or follow each other raising endogeneity problems (Bonsang, 2009). The importance of testing and accounting for potential endogeneity of informal care in models explaining the use of formal care has been explicitly investigated in several studies (Bolin et al., 2008, Bonsang, 2009, Charles and Sevak, 2005, Hanaoka and Norton, 2008, Van Houtven and Norton, 2004, Van Houtven and Norton, 2008). The results of these studies suggest that endogeneity of informal care is likely to be present for two reasons: simultaneity and omitted variable bias. The former reflects the simultaneity of the decision to provide informal care to elderly and to use formal care (Hanaoka and Norton, 2008, Van Houtven and Norton, 2004, Van Houtven and Norton, 2008). The omitted variable bias is relevant since it raises an issue of unobserved individual heterogeneity: for instance there may be unobserved health characteristics or preferences for care that create positive spurious correlation between formal and informal care by increasing the demand for both (Charles and Sevak, 2005).

Among studies that control for endogeneity of informal care, the majority uses cross sectional data and rely on instrumental variable regression (Bolin et al., 2008, Bonsang, 2009, Gannon and Davin, 2010, Hanaoka and Norton, 2008). One study applied the sample selection model proposed by Heckman and extended by Bourguignon et al., in 2007 (Jimenez-Martin and Prieto, 2011). On the other hand, three studies investigated the link between informal care and other types of care with longitudinal (panel) data (Charles and Sevak, 2005, Van Houtven and Norton, 2004, Van Houtven and Norton, 2008).

Our study extends the previous findings by investigating the impact of informal care on the utilization of health services while focussing on one specific disease. It provides new evidence on the role of informal care by using a panel data from a representative sample of stroke patients in Italy. More specifically, we investigate whether the presence of potential caregiver and amount of actual informal care has any influence on the use of health services after stroke. In addition, it represents the first analysis of informal care role in access to and use of rehabilitation services during stroke recovery. Our paper aims to contribute to the existing evidence on the complex interplay between informal and formal health care and to offer a novel contribution to the literature from both the policy and methodological point of view.

From a policy perspective, given the significant and raising socio-economic burden of stroke in developed countries, health care of stroke survivors in the community is likely to become an increasing priority over the next decade. Understanding the role of informal care – the most significant portion of total societal costs of stroke – is of critical importance for informing health policies for these patients.

From a methodological perspective, our study is one of very few empirical analyses to use panel data for investigating the role of informal care vis-à-vis formal healthcare services. Panel data allow for much better control of endogeneity, essential to correctly measure how informal care affects healthcare utilization, since they exploit the time dimension of the subject to control for the unobserved heterogeneity (Wooldridge, 2010).

The reminder of the paper is organised as follows. In the following section we present the study design, data and describe variables used in the empirical model. Sections 2.3 and 2.4 explain the empirical model and discusses econometric issues. Results are showed in Section 3, followed by the discussion and conclusions.