The brief case for everyday problems: a proposal of two brief alternate forms of the Everyday Problems Test

Abstract

Everyday Problems Test (EPT; Willis and Marsiske, Manual for the everyday problems test, Pennsylvania State University, Pennsylvania, 1993) is an 84-item performance-based measure of older adults' everyday cognitive competencies in seven everyday domains (e.g., finance, reading prescription). Its length makes it disadvantageous in the typical time-constrained testing context. Due to the potential practice effects, it is also impractical for longitudinal and intervention studies which require repetitive testing. We have addressed these issues by adapting two brief forms of EPT, with 14 items each. The psychometric evaluation of these two versions was conducted on a sample of 157 cognitively healthy older adults. Both brief forms demonstrated good internal consistency, high inter-correlation, and have shown satisfactory concurrent criterion-related validity based on their correlations with socio-demographic and cognitive variables. Results indicate that the two proposed brief forms can be a valuable tool in assessing the everyday cognitive competence of healthy older adults either as a one-time screening instrument or as a pretest–posttest difference indicator of the intervention efficacy.

Introduction

Everyday cognitive competence can be defined as “the ability to perform adequately those cognitively complex tasks of daily living considered essential for living on one's own in this society” (p. 595, Willis 1996). A critical aspect of everyday cognitive competence and functional health is the maintenance of independence. Given the so-called graying of the population, as well as the heighted complexity of the demands of mundane living, the everyday independence in aged is an important issue not only for elders themselves, but also for the practitioners and the society. Studies identifying subtle shortfalls in independence and failures to sustain everyday functioning are gaining prominence because identifying the starting points of decremental trajectories could facilitate early diagnosis critical for effective care (van Harten et al. 2018). Everyday functioning is conceptualized as a multidimensional phenomenon involving physical mobility, social factors, and everyday cognitive competence (Peel et al 2004). The changes in the latter are the trickiest to identify because various socio-demographic circumstance can attenuate the changes.

Everyday cognitive competence is rooted in both, domain-specific individual knowledge based on education and experience, as well as skills and abilities enabling the realization of such knowledge. The competence to independently solve everyday tasks, such as managing one’s medication, finances, and household maintaining enables prosperous aging. Age-related decline in everyday cognitive competence is particularly evident in individuals with lower education (Willis 1996; Schaie and Willis 1999), although cognitive factors can mediate its relation to age and education (Borella et al. 2017). Previous research shows that, after controlling for demographic variables (i.e., age, education), various cognitive factors can contribute to everyday cognitive competence of healthy older adults (Chen et al. 2017; Tucker-Drob 2011) and individuals with mild cognitive impairments (McAlister et. al. 2016).

Cognitive abilities and everyday cognitive competence

Cognitive abilities suggested to underlie everyday cognitive competence are processing speed (e.g., Owsley et al. 2002), reasoning (e.g., Allaire and Marsiske 1999), and executive functions (e.g., Tomaszewski Farias et al. 2009). Processing speed has been identified as a strong predictor of everyday cognitive competence when assessed by time-constrained measures (Owsley et al. 2002; Borella et al. 2017). When assessed by time-unconstrained measures processing speed yields a small contribution in predicting everyday cognitive competence due to the low demands on processing speed inherent to this measure (e.g., Burton et al. 2006; Borella et al. 2017).

Reasoning ability is often tied to solving everyday problems and situations, such as following a cooking recipe or managing finances. Indeed, fluid reasoning is usually found to be a consistent predictor of everyday cognitive competence (e.g., Allaire and Marsiske 1999; Gross et al. 2011; Kimber 2013; Borella et al. 2017). On the other hand, crystallized intelligence (operationalized mostly via vocabulary tests) yields somewhat ambiguous results regarding its relation to everyday cognitive competence. Although some studies find a relation between everyday competence and crystallized abilities (e.g., Borella et al. 2017; Chen et al. 2017), others do not confirm this relation suggesting that the general measures of crystallized abilities are not always a good predictor of experienced-based knowledge associated with specific domains inherent to complex daily tasks (e.g., Kimber 2013).

Executive functions are commonly viewed as a general higher-order cognitive ability, yet studies find its three relatively independent subcomponents: inhibition, updating, and task shifting (Miyake et al. 2000). Although studies have found that measures of executive functions are predictive of everyday cognitive competence (e.g., Cahn-Weiner et al. 2002; Tomaszewski Farias et al. 2009; McAlister et. al. 2016), the role of its component processes has been sporadically investigated (Vaughan and Giovanello 2010; McAlister and Schmitter-Edgecombe 2016). Previous studies have shown an important role of updating in predicting everyday cognitive competence, while inhibition and shifting were found to be less significant for everyday functioning (e.g., McAlister and Schmitter-Edgecombe 2016). However, some theories emphasize the role of inhibition as a key component of normative cognitive aging (Hasher and Zacks 1988; Kaye et al. 1990). To a certain extent, inhibition deficits explain age-related deficits in cognitive functions, such as selective attention, working memory, language, spatial, and reasoning abilities (e.g., Cansino et al. 2011; Salthouse 1991). Furthermore, they explain various impairments in everyday functioning in aging, such as understanding speech from background noise (Tun et al. 2002) or ignoring visually distracting information during reading (Li et al. 1998). Consequently, we believed age-related decline in inhibition could have a significant role in understanding individual differences in everyday cognitive competence.

Assessment of everyday cognitive competence

Everyday cognitive competence assessment has been recognized as an important issue in health care settings (Giles et al. 2020). Clinicians and researchers use self- or informant-report questionnaires and performance-based tasks to assess everyday cognitive competence. Yet, precautions should be taken when considering adequate assessment methods because self-reports and performance-based tasks might be capturing different aspects of everyday cognition and competence (Chen et al. 2019; Schmitter-Edgecombe et al. 2011). Self-reports are characterized by the ease of administration and can be advantageous for capturing experiences of older adults in multiple real-world situations. Still, they cannot fully tap the multidimensionality and stratification of everyday cognitive competence in terms of higher-order cognitive abilities and skills fundamental for their execution. Moreover, their results are heavily influenced by the reporting bias (Schmitter-Edgecombe et al. 2020). Performance-based tasks are time-consuming, mostly with no time constraints, yet more objective, and require participants to solve real-life problems using a standardized test format. Instead of self-reporting or reports from relevant others on how one would perform in a specific situation or context (e.g., deciding upon the painkiller medication in case of a toothache), performance-based measures will incorporate the necessity of convergent decision making to perform well (e.g., deciding upon the toothache medication after being presented with the list of painkillers and their symptomatic domains). When investigating subtle intra- and interindividual differences in everyday cognitive competence, it might be beneficial to choose performance-based tasks since they are more sensitive than self-reports (Chen et al. 2019; McAlister and Schmitter-Edgecombe 2016; Tucker-Drob 2011).

Everyday Problems Test (EPT; Willis and Marsiske 1993) is a paper–pencil performance-based measure of older adults' competencies in solving everyday complex and challenging tasks in seven domains representing instrumental activities of daily living (IADL; Lawton and Brody 1969): medication use, meal preparation, telephone use, shopping, financial management, household care, and transportation. The EPT consists of 42 tasks (charts, directions, or forms) relevant to the everyday functioning of older adults. Each task is followed by two open-ended questions, with no time constraints. For example, participants are presented with a regular and an intermittent car service checklist. Based on the presented material, participants should determine which services the car would need at all service checks and at what mileage does the car require the most service (regular and intermittent). The EPT requires the systematic use of analytical and synthetical skills; to correctly respond, one must read, retain, identify, and adequately process the information presented in the everyday form, such as recipes, prescriptions, appliance manuals. Over the years, different forms of EPT, varying in the number of items, have been developed (84-item and 48-item; Willis 1990; 32-item; Willis 1993; 28-item; Willis et al. 2006), as well as recent adaptations to Italian (Borella et al. 2017) and Greek (Pavlidis et al. 2021) cultural setting.

Although EPT is readily available, easy to administer, and has good psychometric properties, using the full scale can be impractical due to time requirements. Additionally, its repeated use in longitudinal context might lead to practice effects (Yam et al. 2014). The benefits of having a brief version, as well as an alternate form, are thus quite straightforward. Face validity of EPT is self-evident, yet visual inspection reveals its sociocultural origin (USA), such as the metrics or currencies used in the tasks. Apart from adapting the EPT for the novel cultural context, the aim of this study was twofold: 1) to evaluate the psychometric properties of the two proposed brief alternate forms of EPT characterized by more appropriate length and time requirements more adequate for middle and older aged adult population and 2) to investigate the role of cognitive factors in predicting everyday cognitive competence after controlling for demographic variables. We expect that participants with higher fluid reasoning would score better on the EPT since they would have more efficient skills to solve novel and various problems (Wills et al., 1992). On top of that, impairment in updating and inhibition could predict poorer EPT scores due to difficulties in selecting and recalling task-relevant information (De Beni and Palladino 2004) and retention of task-irrelevant information (Jost et al. 2011).

Method

Participants

A convenience sample of middle and old-aged participants with normative aging (N = 157, 57.3% female; M_age = 62.6, SD_age = 8.56, TR_age = 50—84), with different educational attainment (2.5% no education, 5.1% elementary school, 39.5% high school, 15.3% BA, and 37.6% MA or higher), took part in the study. The data for seven participants were excluded due to extreme results (N = 5) or identical answers given on the EPT (N = 2).

Material

Mini-Mental State Examination 2: Brief Version (MMSE-2: BV, Folstein et al. 2011) is a cognitive screening measure widely used to detect cognitive impairment. A brief 16-items form with a cutoff score of 11 was administrated. This form assesses mental status via items which include orientation to time, orientation to place, registration, and recall.

Standard Progressive Matrices (SPM; Raven 1994) are an individually administered standard nonverbal fluid intelligence test. SPM consists of problem which involve completing a 3 × 3 pattern/figure by selecting a correct missing part among six given alternatives. A short version of the SPM, featuring nine items (Bilker et al. 2012), was used to avoid fatigue effects.

N-back task (Jaeggi et al. 2010) is a computerized working memory updating task. Participants are presented with a continuous stream of stimuli and must indicate if a current stimulus is the same as the one displayed n instances back. The version of the task used for this study included three 1-back, 2-back, and 3-back blocks. Each block contained 20 + n stimuli out of which six were target stimuli. Pictures used as stimuli were divided into different categories (fruits, animals, body parts, means of transport, furniture, clothing), and participants could choose a category to be displayed. Each stimulus is displayed for 500 ms with a 2500-ms interstimulus interval. The final score was formed as the proportion of hits averaged across all blocks.

Picture-Word Task (Starreveld and La Heij 2017) is a computerized inhibition task, i.e., a modified version of the Stroop task, containing pictures (e.g., tree) with superimposed words, displayed in the center of the screen. A word is either congruent (e.g., tree) or incongruent (e.g., ball) with the picture. Participants’ task is to name the picture by pressing the corresponding key while ignoring the meaning of the distractor word. The task included 240 stimuli divided into three blocks, containing either 25%, 50%, or 75% of congruent items, respectively. The final score was the difference in reaction time for incongruent and confluent items for the correct trials.

Everyday Problems Test (EPT; Willis and Marsiske 1993), a measure of complex everyday problems solving, was cross-translated and adapted. The adaptation of the EPT was carried out in several steps. In the first step, we translated the entire EPT into Croatian. The imperial system of units used in the original EPT items has been replaced by meters (distance), kilograms (weight), as were degrees Fahrenheit to Celsius (temperature) and US dollars to Croatian Kuna (currency). Names, institutions, telephone numbers, and health services have been adapted to Croatian context. We have adjusted the dates to match the current ones. Two translators have independently translated and back-translated the test. Any disagreements among these translations were resolved in agreement with a fellow-psychologist. In the second step, a pilot study was conducted in which the EPT was used in a smaller sample (N = 15) and participants were requested to write down the comments related to the ambiguity of the items. In the third step, these comments were reviewed and the items that remained unclear were modified (e.g., a question that suggested two possible answers was modified to make it clearer what was being asked). The average administration time for the adapted 84-item EPT was M = 131.7 min (SD = 73.33). Cronbach's alpha coefficient was α = . 92. The final score is the sum of correct answers (maximum of 84).

Procedure

A convenience sample was recruited via acquaintances and using the snowball method. All instruments were administrated individually, and in the following order: (1) MMSE-2: BV, (2) SPM, (3) n-back, (4) Picture-Word task, (5) EPT. Computerized tasks (n-back and a modified version of the Stroop task with pictures and words displayed as stimuli task) were run on participants` personal computers using Unity Real-Time Development Platform. Experimenters attended the administration of tasks 1–4, while EPT was taken at participants’ own pace. The study was approved by the Ethical Committees of the research institution, and all participants gave their written informed consent in accordance with the Declaration of Helsinki.

Analysis

A prior power analysis indicated that a sample size of 103 would be sufficient to detect a medium-size effect with a power of 0.80 and an α = 0.05. Outliers (z-score >  ± 3.29) were excluded (Tabachnick and Fidell 2013). Items for brief forms were selected content-wise (each of the seven proposed domains of daily living represented by one task, i.e., two items) and based on the item-total correlations and difficulty indices to preserve content validity, internal consistency, and difficulty. Internal consistency was estimated using Cronbach’s α coefficient. Hierarchical regression analyses using each EPT form as a criterion examined whether cognitive variables introduced in second step explained additional variance after controlling for sociodemographic variables (1st step). Bivariate correlations and hierarchical regression analyses (HRA) were used to demonstrate congruent criterion-related validity of brief forms.

Results

Items for the two brief forms (EPT-A, EPT-B; see Appendix) were selected within each of the seven domains and based on the item-total correlation and item’s difficulty indices (Table 1). Since we were interested in comparing effect sizes between the items, the significance tests for individual coefficients are omitted. Descriptive data for all observed variables are shown in Table 2. The dimensionality of EPT is not routinely assessed since the test is very long and time-consuming, so the item-to-subject ratio is usually not sufficient for CFA (similar to Pavlidis et al. 2021). We have performed an EFA on the full scale using the principal axis method. One significant factor proved to be the most reasonable solution, explaining 15% of total variance. Item communalities varied greatly, from 0.023 to 0.448, indicating low to medium saturation with the common factor. Even though we did not replicate the seven dimensions corresponding to proposed behavioral domains, we decided to choose the items respecting the initial domains to preserve the content validity of the scale.

Table 1.

Item-total correlations and difficulty indices for all EPT items (N = 157)

Item	ITC	Difficulty	Form	Item	ITC	Difficulty	Form	Item	ITC	Difficulty	Form
1	0.480	0.68		29	0.314	0.94		57	0.360	0.71	A
2	0.390	0.80		30	0.317	0.59		58	0.182	0.38	A
3	0.413	0.68	B	31	0.401	0.80		59	0.329	0.87
4	0.358	0.65	B	32	0.379	0.67		60	0.261	0.53
5	0.577	0.66	A	33	0.273	0.94		61	0.292	0.49
6	0.541	0.68	A	34	0.333	0.48		62	0.399	0.48
7	0.264	0.90		35	0.207	0.96		63	0.399	0.85	A
8	0.404	0.76		36	0.449	0.89		64	0.504	0.83	A
9	0.290	0.50		37	0.374	0.96		65	0.456	0.56
10	0.295	0.66		38	0.359	0.73		66	0.458	0.51
11	0.352	0.76		39	0.492	0.76	B	67	0.324	0.71
12	0.387	0.69		40	0.328	0.71	B	68	0.435	0.90
13	0.338	0.85		41	0.561	0.64	A	69	0.377	0.95
14	0.363	0.57		42	0.540	0.80	A	70	0.413	0.92
15	0.397	0.93		43	0.361	0.68	B	71	0.342	0.85
16	0.200	0.76		44	0.528	0.65	B	72	0.421	0.42
17	0.311	0.87		45	0.433	0.81		73	0.491	0.89	A
18	0.365	0.71		46	0.416	0.78		74	0.390	0.28	A
19	0.498	0.64	B	47	0.332	0.97		75	0.224	0.89
20	0.528	0.48	B	48	0.234	0.88		76	0.418	0.93
21	0.346	0.59		49	0.320	0.66		77	0.450	0.70	B
22	0.219	0.84		50	0.259	0.17		78	0.638	0.86	B
23	0.370	0.61	B	51	0.413	0.87	A	79	0.468	0.68
24	0.538	0.50	B	52	0.587	0.82	A	80	0.468	0.56
25	0.387	0.87	A	53	0.173	0.91		81	0.293	0.96
26	0.348	0.66	A	54	0.384	0.90		82	0.362	0.93
27	0.201	0.15		55	0.276	0.80	B	83	0.342	0.66
28	0.209	0.64		56	0.201	0.86	B	84	0.518	0.89

1–84 Everyday Problems Test (items 1 through 84); ITC item-total correlation

Table 2.

Descriptive statistics for observed variables (N = 157)

	M	SD	Min	Max
EPT	60.89	12.79	20	80
EPT-A	9.93	2.85	1	14
EPT-B	9.85	3.01	2	14
MMSE-BV	15.23	0.97	11	16
SPM	5.14	2.29	0	9,00
n-back	0.79	0.08	0.59	0.98
Picture-Word task	− 166.11	177.34	− 728.14	277.00

EPT Everyday Problems test, full form (84-item); EPT-A, EPT-B EPT brief forms (14-item); MMSE-BV Mini-Mental State Examination 2: Brief Version; SPM Standard Progressive Matrices

The average item-total correlations of the 84-item full-form EPT and both brief forms were of correspondent values (r_EPT= 0.38, r_EPT-A = 0.45, r_EPT-B = 0.43). Cronbach's alpha coefficients for both brief forms were satisfactory (α_EPT-A = 0.75, α_EPT-B = 0.74; Tavakol and Dennick 2011). EPT-A and EPT-B had similar descriptive parameters (M_EPT-A = 9.93; M_EPT-B = 9.85, SD_EPT-A = 2.85; SD_EPT-B = 3.01), indicating difficulty levels (p_EPT-A = 0.71; p_EPT-B = 0.69) comparable to the full form (p = 0.72). Three two-way ANOVAs were conducted to examine gender (male, female), age (3 subgroups: 50–59, 60–60, 70–84), and educational (two subgroups: high school, BA/MA; results of participants with low educational levels (no education and elementary school) were not included due to their low number, n = 12), effects on performance on both EPT forms. A two-way ANOVA was performed to examine effect of gender on results of two EPT brief forms and revealed that there was no statically significant main effect of neither gender (F (1/155) = 0.001; p > 0.05), EPT form (F (1/155) = 0.384; p > 0.05) nor interaction (F (1/155) = 0.300; p > 0.05). Another two-way ANOVA compared three age groups on both EPT forms. Results showed that younger participants outperformed older (F (2/154) = 16.247; p < 0.001; η_p² = 0.174). The main effect of EPT form (F (1/154) = 0.276; p > 0.05) and interaction (F (2/154) = 1.066; p > 0.05) was not statically significant. A third ANOVA examined the effect of educational level. Participants with higher education level (BA/MA) performed better than participants with high school (F (1/143) = 13.520; p < 0.001; η_p² = 0.086), but main effect of EPT form (F (1/143) = 0.668; p > 0.05) and interaction (F (1/143) = 3.088; p > 0.05) was not statically significant.

The full form of EPT and both proposed brief forms (EPT-A, EPT-B) have similar intercorrelations with other investigated variables (Table 3). As expected, negative correlations were found with age, while positive correlation was found with educational attainment and cognitive variables (overall cognitive status, fluid reasoning, updating, and inhibition ability).

Table 3.

Intercorrelations among the variables of interest (N = 157)

	1	2	3	4	5	6	7	8	9
1 EPT	–
2 EPT-A	0.91***	–
3 EPT-B	0.894***	0.792**	–
4 Age	− 0.403**	– 0.427***	− 0.447**	-
5 Gender	− 0.030	− 0.012	0.016	0.016	–
6 Education	0.292***	0.253**	0.302***	− 0.180*	− 0.086	–
7 MMSE-BV	0.160*	0.165*	0.205*	− 0.188*	0.031	0.189*	–
8 SPM	0.511***	0.489***	0.498***	− 0.424***	− 0.048	0.272**	0.129	−
9 N-back	0.371***	0.410***	0.429***	− 0.445***	− 0.161*	0.214**	0.095	0.372***	–
10 Picture-Word task	0.163*	0.177*	0.184*	− 0.168*	− 0.179*	0.061	− 0.079	0.064	0.278***

EPT Everyday Problems Test, full form (84-item); EPT-A/EPT-B Everyday Problems Test, brief forms (14-item); MMSE-BV Mini-Mental State Examination 2: Brief Version; SPM Standard Progressive Matrices

∗∗∗p < .001, ∗∗p < .01, ∗p < .05, gender coding = male (1), female (2)

Three separate hierarchical regression analyses were conducted, each using a different EPT form as a criterion, with sociodemographic and cognitive variables as predictors (Table 4). In all these analyses, age and education were found as significant predictors (step 1) and cognitive variables included in step 2 have explained an additional ~ 13% of the criterion variance. Fluid reasoning (SPM) was the strongest among cognitive predictors. N-back and age were predictive of brief forms of EPT, but not of full form. In the 2nd step, education level ceased to be a significant predictor.

Table 4.

Summary of the three hierarchical regression analyses for variables predicting EPT scores (N = 157)

Criterion	EPT		EPT-A		EPT-B
	1st step (β)	2nd step (β)	1st step (β)	2nd step (β)	1st step (β)	2nd step (β)
Age	− 0.362***	− 0.151	− 0.394***	− 0.173*	− 0.406***	− 0.179*
Gender	− 0.004	0.034	0.01	0.055	0.043	0.090
Education level	0.226**	0.130	0.183*	0.086	0.232**	0.131
MMSE-BV		0.055		0.065		0.094
SPM		0.356***		0.315***		0.303***
N-back		0.121		0.176*		0.187*
Picture-Word task		0.084		0.089		0.098
R	0.461	0.589	0.463	0.589	0.502	0.624
R2	0.212	0.347	0.214	0.317	0.252	0.389
ΔR2	0.212	0.135	0.214	0.137	0.252	0.137
ΔF	13.74***	7.68***	13.89***	7.61***	17.19***	8.37***

EPT Everyday Problems Test, full form (84-item); EPT-A, EPT-B EPT brief forms (14-item); MMSE-BV Mini-Mental State Examination 2: Brief Version; SPM Standard Progressive Matrices

∗∗∗p < .001, ∗∗p < .01, ∗p < .05, gender coding = male (1), female (2)

Discussion

The two proposed brief forms of EPT are culturally adapted and feature all domains of activities of daily living represented in the full form; thus, the face and content validity were preserved, and the internal reliability of the forms was satisfactory. The overall difficulty indices of brief forms (p_EPT-A = 0.71, p_EPT-B = 0.69) were slightly higher than the difficulty of the original full-form version (p = 0.62; Willis and Marsiske 1993) or its 28-item Greek adaptation (p = 0.62; Pavlidis et al. 2021) and were similar to the one observed for our sample with the full instrument (p = 0.72). In accordance with internal constituencies coefficients, EPT-A and EPT-B scores are strongly correlated (r = 0.79). The internal reliability of both brief forms (α_EPT-A = 0.75, α_EPT-B = 0.74) is acceptable in comparison with the reliability of the original 84-item version (α = 0.90; Willis 1996) and 42-item Italian (α = 0.75; Borella et al. 2017) and 28-item Greek adaptation (α = 0.84; Pavlidis et al. 2021). Our results also indicate that both abbreviated forms of the EPT tap individual differences in everyday cognitive competencies in a comparable manner, suggesting their suitability as a measurement tool in a broad range of aging studies.

Correlations of both, EPT-A and EPT-B, with age, education, and cognitive variables indicate adequate criterion-related concurrent validity. Younger older adults consistently show higher scores on performance-based measures of everyday cognitive competencies (e.g., Willis and Marsiske 1993; Chen et al. 2017; Pavlidis et al. 2021), supporting performance-based measures as a valuable tool for assessing and understanding age-related changes in everyday cognitive competencies of cognitively healthy older adults. Comparable negative correlations of age with fluid reasoning (SPM) and everyday cognitive competence (EPT) are in line with the previous studies reporting similar age-related decline for reasoning and everyday problem-solving (Allaire and Marsiske 1999; Tucker-Drob 2011).

The contribution of cognitive factors to everyday cognitive competence depends on the performance-based measure used (Borella et al 2017). EPT measures older adults’ ability to reason and solve everyday problems (Willis and Marsiske 1993). Corroborating on this, fluid reasoning as a predictor had the highest independent contribution in hierarchical regression analyses. Previous research has shown that fluid abilities consistently predict everyday cognitive competence in different age groups (Chen et al. 2017). MMSE 2: BV did not predict EPT score in the 2nd step of hierarchical regression analyses probably due to the restricted scores’ variability since the sample consisted of only cognitively healthy older adults. Studies relating everyday cognitive competence to basic cognitive abilities usually assess executive functions as a general and broad ability (McAlister and Schmitter-Edgecombe 2016; Borella et al. 2017). In this study, executive functions were operationalized by types of executive functioning considered to be affected by aging—updating and inhibition ability (Miyake et. al, 2000). Updating ability (measured by the n-back task) had an independent contribution for both brief forms, but not for the EPT, while inhibition yielded no significant contribution. Corroborating on this, McAlister and Schmitter-Edgecombe (2016) showed that updating ability, operationalized through the Keep Track Task, was the only subcomponent within the array of executive functions which predicted everyday cognitive competencies measured with the 28-item EPT. Due to the overall low difficulty of EPT for cognitively healthy older adults, it is possible that, when averaged, tasks do not exhibit high enough demands on the updating ability, thus yielding this ability as an insignificant predictor for the full-form EPT. The observed low bivariate correlation between inhibition and EPT results that failed to reach significance in HRAs, when all cognitive variables were included in analyses. This is not in line with the inhibition-deficit hypothesis (Hasher and Zacks 1988), which relates a decrease in inhibition to a decrease in other cognitive abilities important for everyday functioning. However, recent meta-analysis questions this hypothesis (Rey-Mermet and Gade 2018) by showing no age-related decline in general inhibition ability and only a decline in the ability to suppress dominant responses. Picture-word task used in our study measures the ability to ignore response interference which makes it less sensitive in capturing age-related changes in inhibition. Therefore, our results call for further investigation of the role of different aspects of inhibition in the impairment of cognitive and everyday functioning in aging.

On the average, it took our participants 132 min to complete the 84-item EPT, which corresponds well with the reported completion time for the 28-item Greek version (45.7 min; Pavlidis et al. 2021). Time-consuming testing procedures are often too demanding for older participants and might lead to inaccurate conclusions about their abilities, most likely in the direction of their underestimation (Segal et al. 1998). On the other hand, administrating a briefer form could sustain participants’ engagement and make EPT a suitable companion to various test batteries taping older adults’ functioning. Here proposed brief forms decrease completion time to approximately 25 min and additionally reduce administrative costs. Considering age-related decrease in attentional efficiency (Commodari and Guarnera 2008) this timeframe represents a significant improvement and acceptable length.

In addition, previous longitudinal studies have reported gains in EPT results over repeated testing (Yam et al. 2014). The use of the two proposed brief forms as alternate versions could overcome this disadvantage and reduce the reported practice effect, but further longitudinal examinations need to establish test–retest reliability and measurement stability of both forms. For example, cognitive training studies focused on older adults often seek to establish transfer effects within the broader spectrum of (more or less) related cognitive abilities than those which have been directly trained—the so-called near- and far-transfer effects (Lustig et al. 2009). The goal of such cognitive interventions is to find evidence of transfer in everyday life, i.e., everyday cognitive competencies. Although such a goal is a long way off and it is questionable whether laboratory-based or multimodal interventions could exhibit such far-reaching aims, measures taping everyday functioning need to be established and adequately used within the validation studies. Mixed results of a few studies investigating the benefits of cognitive training on performance-based measures of everyday functioning call for further investigation of transfer of training-related gains to everyday cognitive competences and for the refinement of assessment of everyday functioning (Kelly, et al. 2014).

Limitations of this study are inherent to the sample characteristics. Sample size did not allow for cross-validation or CFA, and future studies should assess the dimensionality of EPT and the proposed brief forms using CFA. Further evaluation of brief forms should also assess test–retest reliability to demonstrate evidence for a conceivable reduction in practice effect. Here reported results reflect the functioning of (cognitively) healthy older adults and caution is advised should EPT-A and EPT-B be used with individuals with mental health issues (e.g., depression), mild cognitive impairments, or older adults with dementia. Hence our results indicate that EPT-A and EPT-B could be a convenient measuring tools in the research context. Additional psychometrical validation and standardization are needed to use absolute performance as an indicator of everyday cognitive competence impairment in a clinical setting. Practical limitations in executing this study, related to time demands, have prevented us from incorporating crystallized intelligence as a predictor of EPT results. As noted by Chen et al. (2017), older adults strongly rely on crystallized abilities when solving everyday problems; thus, a comparison of its predictive role for high- and low-educated older adults would be beneficial. Further research should also endorse using multiple indicators of each executive functioning subcomponent and latent factors because of single-task impurity (Miyake et al. 2000). Further adaptation is also advised for the use in novel cultural settings. On a more general note, future directions call for a more contemporary approach in the assessment of everyday functioning, via computer-based procedures (e.g., Czaja et al. 2017) and by updating of the task content to better reflect the ever-changing requirements of everyday life, for example, by including Internet search for specific information and divided attention conditions within the set of everyday cognitive competences tasks.

In conclusion, everyday cognitive competency is a construct of interest in many disciplines, including the law, gerontology, psychology, and medicine, and due to the changing age structure in our society, it is a construct of rising importance. The here proposed brief forms of EPT represent an adequate and time-efficient tool for assessment of everyday cognitive competence in aging individuals. Researchers and practitioners can make use of the advantages provided by brief forms of EPT either in the time-constrained screening context or by capitalizing on the alternate version in longitudinal studies and intervention evaluation.

Appendix 1

The order of the items in the original questionnaire (EPT original) and the abbreviated version of the questionnaire (EPT short) and the domain (scale) they tap.

EPT-A			EPT-B
Item EPT short	Item EPT original	Domain (scale)	Item EPT short	Item EPT original	Domain (scale)
1, 2	73, 74	Health (medications)	1, 2	3, 4	Health (medications)
3, 4	5, 6	Phone	3, 4	19, 20	Phone
5, 6	51, 52	Financial Management	5, 6	23, 24	Financial Management
7, 8	25, 26	Household	7, 8	39, 40	Household
9, 10	57, 58	Meal preparation (food)	9, 10	43, 44	Meal preparation (food)
11, 12	41, 42	Transportation	11, 12	55, 56	Transportation
13, 14	63, 64	Consumer (shopping)	13, 14	77, 78	Consumer (shopping)

Funding

Funding provided by Croatian Science Foundation (grant number: IP-2020-02-6883).