The management of elderly patients with multiple morbidity is complex and represents one of the most important challenges for the healthcare system. The care of these patients involves great clinical uncertainty. Given that care protocols usually address a single health issue, they are less useful in this setting, and thus tools are needed to facilitate decision-making1. This situation is common in nursing homes for the elderly or in socio-health centres.

Frailty is a multidimensional clinical entity defined as a state of vulnerability to everyday stressors due to limiting compensatory mechanisms that puts the individual at high risk of poor health outcomes2,3. As a continuous variable, a Frailty Index (FI) based on the accumulation of deficits model with different domains can measure the individuals’ degree of vulnerability and therefore their biological age2. In this sense, FIs could be defined as quantified comprehensive geriatric assessment (CGA)2.

The objective of this study was to describe a first approach to the development of an FI based on CGA within the social and health care setting (i.e., nursing homes), and to conduct a first analysis of the potential of this tool to help healthcare professionals make decisions for individualized drug prescription based on the actual status of the patient.

The development of the Frailty Index for Nursing Homes (Spanish acronym: IF-CSS) was based on data provided by the recently published FRAGIL-VIG study2 and on other similar proposals aimed at defining a prognostic index4. The development process comprised 2 stages: a) The construction of the IF-CSS, which included selecting the variables and determining their weights; and b) A preliminary analysis, which comprised the administration of the IF-CSS to patients at 2 nursing homes.

As a continuous variable, the FI is a useful tool to assess overall patient status. It could therefore be useful in individualizing pharmacological treatment by determining the level of therapeutic intensity to be implemented.

The literature recommends the inclusion of 30 to 40 deficits in these kinds of indices to maintain their predictive capacity. However, in this first approach, we only included 27 deficits grouped into 17 variables commonly included in the CGA administered in nursing homes to facilitate the implementation of the index in caregiving practice. One of the objectives was that the index should be quick and simple to administer.

The IF-CSS fulfilled the criteria established for the selection of the variables and preserved the weight of each domain in the final result. The geriatric syndrome domain had the greatest weight and was slightly greater than that described in other FI proposals. This aspect may be because such deficits are followed-up with increased frequency in nursing homes due to their impact on the patient's quality of life and the level of care required. Most of the variables included were measured using indices by assigning a value according to the score obtained. Dichotomous scoring (0-1) was applied to variables with less diversity. For example, it is simple to determine if a patient has dysphagia or not (i.e., a dichotomous variable), but variables such as cognitive or functional impairment require a greater number of deficits to obtain a more accurate score. This approach to scoring also allowed us to balance the weight of each domain and to maintain the percentages proposed by other authors2,3,5. In the emotional domain (i.e., the GDS scale), the use of antidepressant treatment (Table 1) was assigned a positive score to avoid assigning a negative score for the absence of depression due to the treatment itself or due to its not being assessable in patients with advanced cognitive impairment.

The final selection of variables differed from that of other proposals. For example, we did not include functional capacity as measured by the instrumental activities of daily living (IADL), because the vast majority of people in nursing homes are fully dependent at this level. Neither was the social domain included, because it is understood that the social needs of institutionalized patients are almost completely covered by the center, and thus there would be few differences between patients.

Despite this study being a first approach, the results show that the IF-CSS has discriminative capacity regarding the level of frailty. As an example of its potential usefulness, the results of the IF-CSS suggest a therapeutic approach addressing functional maintenance in the 35% of patients with moderate frailty, or a more conservative approach mainly addressing symptomatic or palliative control in the 25% of patients with advanced frailty.

This first approach is being developed in two directions. On the one hand, we are going to assess the inclusion of some of the deficits which were excluded in this first proposal. These deficits are included in the CGA and we consider that they can have an impact on patient vulnerability. Thus, we will include: a) the variable family support for the institutionalized person (i.e. within the social domain); b) the variable moderate-severe pain as a symptom of severity; and c) the variable hospital admission in the previous 6 months. On the other hand, we will attempt to validate the predictive capacity of the FI for mortality. This will be achieved by incorporating the IF-CSS into the CGA routine administered every 6 months to patients in nursing homes in which pharmaceutical care forms part of the work of the interdisciplinary team.

No funding declared.

No conflicts of interest declared.