Doctoral dissertation

Hanna Kerminen: Geriatric Assessment in Clinical Practice : Current Situation and Challenges in Implementation

Hanna Kerminen
Comprehensive geriatric assessment (CGA) is a central part of geriatric medicine. However, the concept of CGA is not well established, and the use of the term CGA is incoherent and unclear in both research and clinical practice. There is also a paucity of research considering CGA use in daily clinical practice. Med.lic., specialist Hanna Kerminen studied in her doctoral dissertation the use and utilization of CGA.

Objectives. Study I of the dissertation aimed to collect data on the current situation of the use of CGA in clinical practice in Finland. In Studies II and III, the objective was to clarify how data acquired from a widely used geriatric assessment instrument (interRAI) may be utilised to detect hospitalised patients with an increased risk of adverse hospital outcomes. Study II aimed to construct a frailty index (FI) and analyse its association with hospital outcomes. In Study III, the objective was to identify readmission predictors among patients discharged from geriatric hospitals. Study IV aimed to gain insights on the challenges of the geriatric assessment implementation process by describing the preliminary results of a depression screening protocol implemented among respiratory insufficiency patients at a pulmonary outpatient clinic in a tertiary hospital.

Materials and methods. Study I involved a web-based questionnaire survey about CGA use among 95 geriatrician members of the Finnish Geriatrics Society. The evaluated domains were the assessment of cognition, assessment of nutrition and functional ability, evaluation of depression, and measurement of orthostatic blood pressure.

Studies II and III were retrospective cohort studies of patients aged ≥70 years hospitalised in two geriatric hospitals over 3 years. These studies used data from interRAI-Post Acute Care (interRAI-PAC) assessments combined with hospital discharge records. Study II included 2,188 hospitalised patients and Study III included 1,167 patients discharged to home from the index hospitalisation period.

The FI was derived from interRAI-PAC data. The associations of interRAI-PAC scales and FI with hospital outcomes were analysed. Hospital outcomes included in-hospital mortality, prolonged hospital stay, and emergency department admission. Study III investigated the associations of interRAI-PAC variables and scales with 90-day readmission of the patients.

Study IV was a retrospective evaluation of the outcomes of a depression screening protocol using the records of 238 patients. In the protocol, the patients completed the Depression Scale (DEPS) questionnaire. Patients whose scores were indicative of depression were offered the opportunity to further undergo an assessment of mood at a psychiatric outpatient clinic.

Results. Study I: The majority of geriatricians involved in the study (94%) used CGA, but a minority (38%) administered it to all new patients (response rate 49%). Ten respondents (11%) incorporated all five domains into the assessment, whereas others selected domains according to their clinical judgement.

Study II: The discriminative ability of the FI for in-hospital mortality (area under the curve [AUC] 0.73) and prolonged hospital stay (AUC 0.75) was good. However, the short hierarchical scale for the activities of daily living (ADLH) was as good as the FI in predicting these outcomes. All tested instruments were poor at predicting emergency department admission.

Study III: The risk factors associated with readmission in univariate analysis were age, admission from home (vs. acute hospital admission), Alzheimer’s disease, unsteady gait, fatigue, unstable condition, ADL impairment, body mass index (BMI), FI, bowel incontinence, hearing difficulties, and poor self-rated health. In multivariate analysis, age, ADL impairment, and BMI persisted as risk factors.

Study IV: The DEPS was administered to 66% of the patients in the first year of screening, but the coverage increased to 88% in the second year. Of the patients, 34% (n=21) scored ≥9 points, thus exceeding the cut-off for referral. Only 13 patients were referred, as the remainder declined the referral. Finally, seven patients were evaluated at a psychiatric outpatient clinic, and all were deemed to have depression.

Conclusions. Most Finnish geriatricians used CGA, but CGA use was not systematic, and the content of CGA was variable. This type of incomplete evaluation may lead to inadequate detection of geriatric syndromes. It was possible to derive the FI from interRAI-PAC data, and this FI predicted adverse hospital outcomes as expected. However, its predictive ability was not better than that of the short ADLH scale.

In clinical practice, assessment of ADL is a simple and valid way to evaluate a patient’s prognosis. interRAI-PAC evaluation performed upon admission to geriatric hospitals revealed patient-related risk factors for readmission. Based on the identified risk factors, we recommend that the patient’s functional ability, ADL needs, and individual factors underlying ADL impairment as well as nutritional and mobility problems should be carefully addressed and managed during hospitalisation to diminish the risk for readmission.

Depression screening improved the detection of depressive symptoms, but its effect on the patients’ treatment and clinical courses was small. Rather than referring patients to a psychiatric unit, the evaluation and management of depression should be undertaken at a same unit where a screening is performed.

The doctoral dissertation of Med. lic., specialist Hanna Kerminen in the field of medicine titled Geriatric Assessment in Clinical Practice – Current Situation and Challenges in Implementation will be publicly examined at the Faculty fo Medicine and Health technology of Tampere University at 12 o'clock on Friday 26 March. Professor Kaisu Pitkälä from University of Helsinki will be the opponent while Professor, tenure track Esa Jämsen will act as the custos.

The event can be followed via remote connection. Meeting ID: 673 3387 2258.

The dissertation is avalable online at

Photo: Pasi Kerminen


Upcoming events

Past events