Abstract Background Greater numbers of older patients are accessing hospital services. Specialist geriatric input at presentation may improve outcomes for at-risk patients. The Survey of Health, Ageing and Retirement in Europe Frailty Instrument (SHARE-FI) frailty measure, developed for use in the community, has also been used in the emergency department (ED). Aim To measure frailty, review its prevalence in older patients presenting to ED and compare characteristics and outcomes of frail patients with their non-frail counterparts. Design Patient characteristics were recorded using symphony® electronic data systems. SHARE-FI assessed frailty. Cognition, delirium and 6 and 12 months outcomes were reviewed. Methods A prospective cohort study was completed of those aged ≥70 presenting to ED over 24 h, 7 days a week. Results Almost half of 198 participants (46.7%, 93/198) were classified as frail, but this was not associated with a significant difference in mortality rates (OR 0.89, 95% CI 0.58–1.38, P = 0.614) or being alive at home at 12 months (OR 1.07, 95% CI 0.72–1.57, P = 0.745). Older patients were more likely to die (OR 2.34, 95% CI 1.30–4.21, P = 0.004) and less likely to be alive at home at 12 months (OR 0.49, 95% CI 0.23–0.83, P = 0.009). Patients with dementia (OR 0.24, P = 0.005) and on ≥5 medications (OR 0.37, 95% CI 0.16–0.87, P = 0.022) had a lower likelihood of being alive at home at 12 months. Conclusions Almost half of the sample cohort was frail. Older age was a better predictor of adverse outcomes than frailty as categorized by the SHARE-FI. SHARE-FI has limited predictability when used as a frailty screening instrument in the ED. Background Greater numbers of older patients are accessing acute hospital services. Those aged 65 years and older account for up to 20% of unscheduled attendances and 40–50% of medical admissions to hospital.1 As the population ages, this cohort of patients is likely to represent an increasing proportion of the acute hospital workload. It has been shown that older patients have poorer outcomes than their younger counterparts.2 Those aged 65 years and older are more likely to have a severe illness and more likely to be admitted to hospital.2,3 Those who have access to comprehensive geriatric assessment as an inpatient have been shown to have improved outcomes.4 Therefore, older patients with a high risk of poorer outcomes may benefit most from specialist geriatric input at an early stage of their hospital attendance. Frailty is a syndrome characterized by reduced functional reserve resulting from a cumulative decline across systems, and it increases an individual’s risk of an adverse outcome when exposed to a stressor.5,6 Its prevalence increases with age, varying between 4% and 59%, depending on the definition used.7,8 There are multiple frailty instruments used both clinically and in research though studies have shown limitations with frailty instruments used to assess patients aged 70 years and older in the acute hospital setting.9,10 Older frail patients often access hospital through the emergency department (ED). Traditionally, the ED was developed to address trauma and acute critical illness, and many departments continue to be organized and operate in this way, though it is not the most appropriate model to address the complex needs of older patients. It is now recognized that creating pathways and screening instruments for use in the acute setting, including the ED, may assist in capturing the complexity of older patients and directing appropriate management. However, screening instruments created for use in the ED such as the Identification of Seniors at Risk tool, Triage Risk Screening Tool and Variables Indicative of Placement risk have not been shown to accurately predict risk.11,12 A frailty assessment tool that can be administered quickly by those working in the ED and easily incoporated into the routine assessment of patients may be of use in identifying older patients most at risk. The Survey of Health, Ageing and Retirement in Europe Frailty Instrument (SHARE-FI) was initially developed for use in the community13 but has since been studied identifying frail older patients in the ED and shown to be of use in predicting adverse outcomes.14 Frail patients have been shown to be almost twice as likely to be admitted to hospital than non-frail.15 Frail patients were also more likely to have an adverse outcome in the 30 days following ED attendance, including functional decline, re-attendance, nursing home admission and death.14 Aims This study aimed to measure frailty using the SHARE-FI, review its prevalence in older patients attending ED and to compare the characteristics and outcomes of frail older patients with their non-frail counterparts. The predictive value of the SHARE-FI instrument in relation to adverse outcomes was also assessed. Design A prospective cohort study was carried out at a 600 bed university teaching hospital. Pre-specified convenience sampling was carried out 7 days a week, with patients recruited over 24 h. Those aged 70 years and older presenting to the ED were included. Patients were recruited between January 2014 and August 2014. Materials and methods Characteristics recorded at the time of index presentation included age, gender, time of attendance (‘out of hours’ attendance was defined as those attending outside of the hours of 09.00–17.00, Monday to Friday), presentation by ambulance, length of stay in the ED and discharge outcomes. This information is recorded routinely as part of the electronic patient record through symphony® electronic data systems. In-patient mortality rates and re-attendance rates were reviewed through electronic records. Mortality rates and the proportion of patients resident in a nursing home at 6 months and at 12 months, were recorded through review of both electronic patient records and contacting patients and informants. The SHARE-FI was used to assess frailty.11 This instrument calculates frailty based on five variables: fatigue, weight loss, weakness as assessed through grip strength, slowness and physical activity. Patients are classified as being frail, pre-frail or non-frail. Cognition was assessed using the Mini Mental State Examination and AD8 scores. Patient was assessed for delirium at presentation using the Abbreviated Mental Test (AMT4) score and Confusion Assessment Method––Intensive Care Unit method. Illness severity was recorded through Manchester Triage Score (MTS) at presentation and Early Warning Score, both of which are recorded routinely at presentation. MTS categories 1 (immediate) to 3 (urgent) are classified as indicating a severe illness. Two investigators assessed patients and obtainted collateral history while in ED. Data were analysed using Stata (Satatcorp). Normally, distributed continuous variables were described as means and SDs and compared across groups using ANOVA/student t-test. Categorical variables were compared using the chi-square test. Sensitivity and specificity of the SHARE-FI in predicting in-hospital and 1 year mortality for frail and non-frail patients was calculated. Binary logistic regression was used to estimate odds ratios for covariates including frailty category and age category (patients were divided into three age groups, from 70–79 years, 80–89 years and 90 years or more) for 12-month outcomes, specifically mortality at 12 months, re-admission within 12 months, nursing home admission at 12 months and a composite outcome of being alive and at home at 12 months. Results Two hundered and twenty patients were recruited. A total of 198 patients had a full assessment completed from January 2014 to August 2014 and were followed up at 6 and 12 months. Of those that did not complete a full assessment 5 had no appropriate collateral available, 10 did not have a full cognitive screen completed in ED, 2 did not complete the SHARE-FI in ED and 5 were lost to follow up at 6 or 12 months. Mean age was 78.8 years, 48.5% (96/198) were male, 51.5% (102/198) were female, 46.7% (93/198) were classified as frail, 20.7% (41/198) pre-frail and 32.3% (64/198) were non-frail. There was no significant difference in the mean age of patients in these three groups (ANOVA: frail 79.1 years, pre-frail 78.8 years, non-frail 78 years, P = 0.518) or in gender [Chi-Square test: frail 44.1% male (41/93), pre-frail 51.2% (21/41), non-frail 53.1% (34/64), P = 0.498; Table 1]. Table 1 Baseline characteristics and initial ED presentation details Non-frail Pre-frail Frail P-value (n = 64) (n = 41) (n = 93) Age (SD), years 78.0 (5.7) 78.8 (6.7) 79.1 (6.4) 0.518 Male (%) 53.1 (34/64) 51.2 (21/41) 44.1 (41/93) 0.498 Presenting ‘out of hours’, (%) 37.5 (24/64) 34.2 (14/41) 44.1 (41/93) 0.497 Arrival by ambulance (%) 31.3 (20/64) 31.7 (13/41) 43.0 (40/93) 0.241 Six hours or less in ED (%) 31.3 (20/64) 26.8 (11/41) 21.5 (20/93) 0.384 Manchester Triage category 1–3 (%) 78.1 (14/64) 68.3 (28/41) 73.1 (68/93) 0.527 Polypharmacy (%) 57.8 (37/64) 70.7 (29/41) 86.0 (80/93) ≤0.001 Delirium (%) 3.1 (2/64) 2.4 (1/41) 15.1 (14/93) 0.009 AMT 4 Score (SD) 3.8 (0.6) 3.7 (0.8) 3.1 (1.1) ≤0.001 Non-frail Pre-frail Frail P-value (n = 64) (n = 41) (n = 93) Age (SD), years 78.0 (5.7) 78.8 (6.7) 79.1 (6.4) 0.518 Male (%) 53.1 (34/64) 51.2 (21/41) 44.1 (41/93) 0.498 Presenting ‘out of hours’, (%) 37.5 (24/64) 34.2 (14/41) 44.1 (41/93) 0.497 Arrival by ambulance (%) 31.3 (20/64) 31.7 (13/41) 43.0 (40/93) 0.241 Six hours or less in ED (%) 31.3 (20/64) 26.8 (11/41) 21.5 (20/93) 0.384 Manchester Triage category 1–3 (%) 78.1 (14/64) 68.3 (28/41) 73.1 (68/93) 0.527 Polypharmacy (%) 57.8 (37/64) 70.7 (29/41) 86.0 (80/93) ≤0.001 Delirium (%) 3.1 (2/64) 2.4 (1/41) 15.1 (14/93) 0.009 AMT 4 Score (SD) 3.8 (0.6) 3.7 (0.8) 3.1 (1.1) ≤0.001 A similar proportion of each frailty category presented with a severe acute illness, classified as a MTS of 1–3 at the time of attendance; 73.1% (68/93) of the frail group, 68.3% (28/41) of the pre-frail group and 78.1% (50/64) of the non-frail group (Chi-Square test: P = 0.527). Similarly, there was no significant difference in the proportion of those arriving by ambulance (Chi-Square test: frail 43.0%, pre-frail 31.7%, non-frail 31.3%, P = 0.241), in those presenting ‘out of hours’ (Chi-Square test: frail 44.1%, pre-frail 34.2%, non-frail 37.5%, P = 0.497) or in those who remained in the ED for more than 6 h prior to admission to hospital or discharge (Chi-Square test: frail 78.5%, pre-frail 73.2%, non-frail 68.7%, P = 0.384). 15.1% (14/93) of frail patients presented with delirium, this was a significantly higher proportion than in the pre-frail [2.4%, (1/41)] and non-frail [3.1% (2/64)] groups (Chi-Square test: P = 0.009). This difference was reflected in AMT 4 scores. There was a higher proportion of polypharmacy (greater than 5 medications prescribed) in frail patients 86.0% (80/93) and pre-frail pateints 70.7% (29/41), compared to the non-frail group 57.8% (37/64), Chi-Square test: P = <0.001 (Table 1). In-hospital mortality rates at the time of initial presentation and admission did not vary significantly between the groups (Chi-Square test: frail 7.5%, pre-frail 12.2%, non-frail 12.5%, P = 0.411) nor did 1 year mortality rates (Chi-Quare test: frail 22.6%, pre-frail 14.6%, non-frail 21.9%, P = 0.556; Table 2). The sensitivity of SHARE FI frailty categories in predicting in hospital and 1 year mortality was 0.46 and 0.47, respectively, and its specificity was 0.61 for in-hospital mortality and 0.59 for 1 year mortality. Table 2 Mortality and re-admission rates Non-frail Pre-frail Frail P-value (n = 64) (n = 41) (n = 93) In-hospital mortality (%) 12.5 (8/64) 12.2 (5/41) 7.5 (7/93) 0.527 Re-admitted within 1 year (%) 57.8 (37/64) 75.6 (31/41) 61.3 (57/93) 0.161 Mean re-admissions within 1 year (SD) 1.2 (1.7) 1.4 (1.4) 1.0 (1.2) 0.411 Mortality at 1 year (%) 21.9 (14/64) 14.6 (6/41) 22.6 (21/93) 0.556 Non-frail Pre-frail Frail P-value (n = 64) (n = 41) (n = 93) In-hospital mortality (%) 12.5 (8/64) 12.2 (5/41) 7.5 (7/93) 0.527 Re-admitted within 1 year (%) 57.8 (37/64) 75.6 (31/41) 61.3 (57/93) 0.161 Mean re-admissions within 1 year (SD) 1.2 (1.7) 1.4 (1.4) 1.0 (1.2) 0.411 Mortality at 1 year (%) 21.9 (14/64) 14.6 (6/41) 22.6 (21/93) 0.556 Using binary logistic regression, it was found that age category predicts 12-month mortality (OR 2.34, 95% CI 1.30–4.21, P = 0.004; Table 3). Younger patients were more likely to be ‘alive and at home’ at 12 months (OR 0.49, 95% CI 0.23–0.83, P = 0.009; Table 4). Patients with a history of dementia (OR 0.24, P = 0.005) and polypharmacy (OR 0.37, 95% CI 0.16–0.87, P = 0.022) were also found to have a lower likelihood of being alive and at home at 12 months (Table 4). Frailty status as defined by SHARE-FI was not associated with a significant difference in either 12-month mortality rates (OR 0.89, 95% CI 0.58–1.38, P = 0.614) or being alive and at home at 12 months (OR 1.07, 95% CI 0.72–1.57, P = 0.745). Neither age category or frailty status was significant predictors of re-admission to hospital or nursing home admission at 12 months. Table 3 Odds ratio for alive at 12 months Variable Odds ratio P-value 95% confidence interval Age ≥ 80 years 2.34 0.004 1.30 – 4.21 Male gender 0.49 0.056 0.24 – 1.02 ‘Out of Hours’ 1.32 0.459 0.63 – 2.78 Ambulance 0.65 0.303 0.29 – 1.47 MTS 1–3 0.94 0.878 0.42 – 2.09 Polypharmacy 2.17 0.111 0.84 – 5.61 Delirium 1.46 0.579 0.39 – 5.49 Hx dementia 1.44 0.511 0.49 – 4.26 Frail by SHARE-FI 0.89 0.614 0.58 – 1.38 Variable Odds ratio P-value 95% confidence interval Age ≥ 80 years 2.34 0.004 1.30 – 4.21 Male gender 0.49 0.056 0.24 – 1.02 ‘Out of Hours’ 1.32 0.459 0.63 – 2.78 Ambulance 0.65 0.303 0.29 – 1.47 MTS 1–3 0.94 0.878 0.42 – 2.09 Polypharmacy 2.17 0.111 0.84 – 5.61 Delirium 1.46 0.579 0.39 – 5.49 Hx dementia 1.44 0.511 0.49 – 4.26 Frail by SHARE-FI 0.89 0.614 0.58 – 1.38 Table 4 Odds ratio for alive and at home at 12 months (composite outcome) Variable Odds ratio P-value 95% confidence Interval Age ≥ 80 years 0.49 0.009 0.28 – 0.83 Male gender 2.05 0.030 1.07 – 3.94 ‘Out of Hours’ 0.84 0.610 0.43 – 1.64 Ambulance 1.33 0.439 0.65 – 2.74 MTS 1–3 0.96 0.917 0.47 – 1.99 Polypharmacy 0.37 0.022 0.16 – 0.87 Delirium 1.05 0.936 0.31 – 3.63 Hx dementia 0.24 0.005 0.72 – 1.57 Frail by SHARE-FI 1.07 0.745 0.72 – 1.57 Variable Odds ratio P-value 95% confidence Interval Age ≥ 80 years 0.49 0.009 0.28 – 0.83 Male gender 2.05 0.030 1.07 – 3.94 ‘Out of Hours’ 0.84 0.610 0.43 – 1.64 Ambulance 1.33 0.439 0.65 – 2.74 MTS 1–3 0.96 0.917 0.47 – 1.99 Polypharmacy 0.37 0.022 0.16 – 0.87 Delirium 1.05 0.936 0.31 – 3.63 Hx dementia 0.24 0.005 0.72 – 1.57 Frail by SHARE-FI 1.07 0.745 0.72 – 1.57 Conclusion The SHARE-FI is a useful tool in the research setting and has the advantage of being easily administered in a clinical environment. It was created using a community dwelling population however, and may not be appropriate for use in the ED.13 Previous studies have examined multiple frailty instruments in patients aged 70 years and older, including the Cardiovascular Health Study index, the Study of Osteoporotic Fractures index and FRAIL index and have also shown them to have poor predictive value in the acute setting.9,10,12 These results were reflected in this study, where increasing age was a significant predictor of being alive and being alive and at home at 12 months, but frailty status was not. Though frailty is a common condition and identifying pre-frailty or frailty at an early stage may be of benefit to patients16 the low sensitivity and specificity of the SHARE-FI in predicting mortality rates in the population examined and the fact that there were no significant differences demonstrated in outcomes noted when frail and non-frail cohorts were compared indicate that it may not be appropriate for use as a screening instrument in this setting. This study found a high prevalence of frailty and pre-frailty in older patients who present to the ED but few significant differences between the baseline and outcome characteristics of these groups and non-frail older patients. The high mortality and re-admission rates seen in patients in this study are likely to be related to a number of factors. A large proportion of this population has multiple co-morbidities and a poor baseline functional status. Over two-thirds of all patients were classified as having a severe illness requiring immediate attention at the time of presentation to hospital (MTS 1–3). The mortality rate at 1 year was greater than 20% in both frail and non-frail groups, further demonstrating a high risk of adverse outcomes in all older patients. This reflects the complexity of this patient cohort, especially when acutely unwell suggesting that an ED attendance for an older person induces frailty almost regardless of pre-morbid baseline. Although comparable to previous studies, limitations of this study include the low numbers of patients when subdivided into frail, pre-frail and non-frail groups and that the SHARE-FI was created and validated to define frailty in a categorical manner as opposed to measuring it as a continous variable.13 The age of patients and the fact that recruitment was from a single centre may also represent a limiting factor. Only patients aged 70 years and older were recruited as this group had previously been shown to have poorer outcomes than their younger counterparts.2,3 Younger patients may also fit the SHARE FI criteria, which was developed in a population aged 50 years and older, and their inclusion in further studies of frailty in the ED should be considered. New methods of assessing patients are being reviewed. In the future, instruments which evaluate functional limitation or functional decline may be of use in the ED and acute hospital.17,18 Such instruments can have the advantage of identifying an individual’s specific care needs. They may be incorporated into the routine assessment of patients and used in a practical way to direct care of individual patients from the time of ED presentation. Current research examining the assessment and management of older patients who access acute hospital services, and an increasing recognition of the limitations of traditional ED models, demonstrates an ongoing interest in the unique needs of older patients. Education and training for staff working with older people both in the ED and throughout their hospital stay, is required. This would ensure their high risk of poorer outcomes is recognized and appropriate management plans are created to minimize this risk. Frailty remains an important concept and should be considered in the assessment of all patients but alternative methods of identifying and quantifying risk to older patients in the ED need to be considered. This study suggests it may be of greater use to consider all older patients as being at high risk of adverse outcomes as hospitalization itself is the critical event regardless of pre-morbid status. Ensuring comprehensive geriatric assessment is available to all relevant older people attending ED requires reorganization of services, resources and in particular training of ED staff to meet this increasing need. Conflict of interest: None declared. References 1 O'Neill S, Courtney G, Carroll J, Geary U, O'Reilly O, O'Connor M, et al. Acute Medicines Programme Report. 2010. Royal College of Physicians of Ireland, Irish Association of Directors of Nursing and Midwifery, Therapy Professions Committee, Quality and Clinical Care Directorate, Health Service Executive. 2 Fallon A, Armstrong J, Coughlan T, Collins DR, O’Neill D, Kennelly SP. Characteristics and outcomes of older patients attending an acute medical assessment unit. Ir Med J 2015; 108: 210– 1. Google Scholar PubMed 3 Kennelly SP, Drumm B, Coughlan T, Collins R, O'Neill D, Romero-Ortuno R. Characteristics and outcomes of older persons attending the emergency department: a retrospective cohort study. QJM 2014; 107: 977– 87. Google Scholar CrossRef Search ADS PubMed 4 Ellis G, Whitehead MA, O’Neill D, Langhorne P, Robinson D. Comprehensive geriatric assessment for older adults admitted to hospital. Cochrane Database Syst Rev 2011; (7): CD006211. DOI: 10.1002/14651858.CD006211.pub2. 5 McCabe JJ, Kennelly SP. Acute care of older patients in the emergency department: strategies to improve patient outcomes. Open Access Emergency Med : OAEM 2015; 7: 45– 54. 6 Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. Lancet 2013; 381: 752– 62. http://dx.doi.org/10.1016/S0140-6736(12)62167-9 CrossRef Search ADS PubMed 7 Collard RM, Boter H, Schoevers RA, Oude Voshaar RC. Prevalence of frailty in community-dwelling older persons: a systematic review. J Am Geriatr Soc 2012; 60: 1487– 92. http://dx.doi.org/10.1111/j.1532-5415.2012.04054.x Google Scholar CrossRef Search ADS PubMed 8 Rockwood K, Andrew M, Mitnitski A. A comparison of two approaches to measuring frailty in elderly people. J Gerontol A Biol Sci Med Sci 2007; 62: 738– 43. http://dx.doi.org/10.1093/gerona/62.7.738 Google Scholar CrossRef Search ADS PubMed 9 Dent E, Chapman I, Howell S, Piantadosi C, Visvanathan R. Frailty and functional decline indices predict poor outcomes in hospitalised older people. Age Ageing 2014; 43: 477– 84. Google Scholar CrossRef Search ADS PubMed 10 Wou F, Gladman JRF, Bradshaw L, Franklin M, Edmans J, Conroy SP. The predictive properties of frailty-rating scales in the acute medical unit. Age Ageing 2013; 42: 776– 81. Google Scholar CrossRef Search ADS PubMed 11 Carpenter CR, Shelton E, Fowler S, Suffoletto B, Platts-Mills TF, Rothman RE, et al. Risk factors and screening instruments to predict adverse outcomes for undifferentiated older emergency department patients: a systematic review and meta-analysis. Acad Emerg Med 2015; 22: 1– 21. Google Scholar CrossRef Search ADS PubMed 12 Elliott A, Hull L, Conroy SP. Frailty identification in the emergency department—a systematic review focussing on feasibility. Age Ageing 2017; 46: 509– 13. Google Scholar CrossRef Search ADS PubMed 13 Romero-Ortuno R, Walsh CD, Lawlor BA, Kenny RA. A frailty instrument for primary care: findings from the survey of health, ageing and retirement in Europe (SHARE). BMC Geriatrics 2010; 10: 57. Google Scholar CrossRef Search ADS PubMed 14 Stiffler KA, Wilber ST, Frey J, McQuown CM, Poland S. Frailty defined by the SHARE Frailty Instrument and adverse outcomes after an ED visit. Am J Emerg Med 2016; 34: 2443– 5. Google Scholar CrossRef Search ADS PubMed 15 Ilinca S, Calciolari S. The patterns of health care utilization by elderly Europeans: frailty and its implications for health systems. Health Serv Res 2015; 50: 305– 20. http://dx.doi.org/10.1111/1475-6773.12211 Google Scholar CrossRef Search ADS PubMed 16 Wilson JMG, Jungner G. Principles and Practice of Screening for Disease . Geneva, WHO, 1968. 17 Hoogerduijn JG, Buurman BM, Korevaar JC, Grobbee DE, de Rooij SE, Schuurmans MJ. The prediction of functional decline in older hospitalised patients. Age Ageing 2012; 41: 381– 7. http://dx.doi.org/10.1093/ageing/afs015 Google Scholar CrossRef Search ADS PubMed 18 Buurman BM, Hoogerduijn JG, van Gemert EA, de Haan RJ, Schuurmans MJ, de Rooij SE. Clinical characteristics and outcomes of hospitalized older patients with distinct risk profiles for functional decline: a prospective cohort study. Thiem U, ed. PLoS ONE 2012; 7: e29621. Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press on behalf of the Association of Physicians. All rights reserved. For Permissions, please email: email@example.com
QJM: An International Journal of Medicine – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera