Age and Ageing · 2006 · 32 citations · 10 references
MedicineGeriatric MedicineGeriatric CareGeriatricsGerontologyComprehensive Geriatric AssessmentFunctional Limitation AssessmentFunctional Limitation MeasurementGeriatric Healthcare UtilizationGeriatric Inpatient CareMedical Records ReviewMedical Records StandardsOlder Adult Care
SIR—Older persons are characterised by age-related changes, multiple diseases, multiple drug use and functional deficits. For optimal care, a holistic approach is needed; however, the health care systems of today are still essentially organised to provide acute medical care to relatively younger populations with little or no co-morbidity [1]. Health systems will have to adapt to this new situation. The value of geriatric assessment has been proven, where targeting is the key to success [2]. With shorter hospital stays, it is of importance to do this targeting quickly and effectively. According to a systematic literature review in the older patients, the most important predictors for adverse outcomes of acute care (mortality, frequent readmissions, institutionalisation and long length of stay) are current illness, decline in physical functions and age. In addition, illness severity, co-morbidity, polypharmacy, cognitive decline, poor nutrition and gender are predictive for one or more of the outcomes [3]. The Minimum Data Set for Acute Care (MDS-AC) instrument was developed to guide care within the hospital and to facilitate the transfer and sharing of information to the next provider of care, thus supporting integrated care. The MDS-AC instrument provides an opportunity to systematically collect information that is reliable on function and co-morbidity and could thus be a valuable addition to the future electronic medical record [4]. The aim of this study is to investigate to what degree important predictors of adverse outcomes, if present according to the MDS-AC instrument during the first 24 h of care for older patients, were not documented in traditional hospital records in acute care wards in five Nordic countries. Hence, the MDS-AC information is assumed to be a gold standard. A secondary aim is to show that suspected deficient documentation is an international issue. The study was carried out in acute medical care in hospitals serving defined communities, one in each of the Nordic countries, with catchment areas of at least 90,000 persons. The study was approved at each study site by the appropriate ethical authority. Informed consent was sought from each patient or his/her nearest relative. This study included 417 patients, 75 years of age and older, in which traditional hospital records were compared with MDS-AC as a part of a Nordic study with 770 participants. The patients were selected randomly from a numbered admission list the morning after admission. The study utilised the MDS-AC, version 1.1, translated into each of the Nordic languages by translators experienced with the translation of InterRAI MDS tools [4]. Patients were assessed within 24 h of admission with the MDS-AC instrument. The data collectors reviewed the hospital records for variables documented during the first 48 h by doctors, nurses and therapists, corresponding to the MDS-AC record variables. The data were collected during the period of January 2001 to April 2002 by a geriatrician and/or experienced geriatric/medical nurses (RN), trained in using the MDS-AC. The study was performed on at least two randomised admissions per weekday but not on weekends or holidays. For additional information on participating hospitals, see Appendix 1, on admission characteristics of patients, see Appendix 2 and for additional information on data collection, see Appendix 3 in the supplementary data on the journal website (http://www.ageing.oxfordjournals.org/). For each of the MDS-AC variables, the prevalence in the medical record of that variable was calculated. Data are presented as impaired or unimpaired on the medical record and MDS-AC assessment, respectively, or unknown if the medical record did not register data found in the MDS-AC assessment. This was performed on the assumption that the existence of impairment would in general be more likely to be documented in the hospital record than if there was no impairment, because there would have been need for help. Sensitivity and specificity (defined in two ways) of the medical record are calculated with MDS-AC documentation at the reference point, thus taking the undocumented cases of the medical record into account [for formulae, see Appendix 4 in the supplementary data on the journal website (http://www.ageing.oxfordjournals.org/)]. The analyses were performed in SAS® version 8.2 for Windows. Results are summarised in Table 1. Average sensitivity for all items summarised in Table 2 was 0.53 and average specificity 0.48. With the assumption that undocumented unimpaired persons are unimpaired, specificity is calculated to be between 0.86 and 0.99. There was some disagreement between the MDS-AC and the hospital records, ranging from 7 to 19%, when the MDS-AC instrument documented impairment but between 0 and 9% for unimpaired. Documentation of activities of daily living (ADL), instrumental activities of daily living (IADL) and several important clinical entities The table summarises comparison between hospital records and the Minimum Data Set for Acute Care instrument. All figures are shown in percentage of impaired and unimpaired, respectively. N = 417. Pain control data are missing for 19 patients with pain. Assumed specificity was calculated with the assumption that undocumented cases of those who were unimpaired were in fact unimpaired, as it is a common practice, rightly or wrongly, to assume that what is not documented is unimpaired. Documentation of activities of daily living (ADL), instrumental activities of daily living (IADL) and several important clinical entities The table summarises comparison between hospital records and the Minimum Data Set for Acute Care instrument. All figures are shown in percentage of impaired and unimpaired, respectively. N = 417. Pain control data are missing for 19 patients with pain. Assumed specificity was calculated with the assumption that undocumented cases of those who were unimpaired were in fact unimpaired, as it is a common practice, rightly or wrongly, to assume that what is not documented is unimpaired. Table 2 shows that the lack of documentation was a relevant issue for all countries. For most of the items, one or two of the departments showed excellent documentation. Variability in documentation between the Nordic study sites Pain control data missing for eight patients with pain. The table summarises comparison between hospital records and the Minimum Data Set for Acute Care (MDS-AC) instrument. The numbers represent the percentage of undocumented variables in the medical records compared with the MDS-AC documentation per country. N = 417. Variability in documentation between the Nordic study sites Pain control data missing for eight patients with pain. The table summarises comparison between hospital records and the Minimum Data Set for Acute Care (MDS-AC) instrument. The numbers represent the percentage of undocumented variables in the medical records compared with the MDS-AC documentation per country. N = 417. This systematic audit of 417 records from older patients in acute medical care shows that important information relevant to the potential outcomes of care and suffering of the patients was missing in the medical records during first 48 h of care. The under-documentation of co-morbid and functional impairments in the hospital record, compared with the MDS-AC instrument, ranges from 13 to 74%. If the hospital record were a diagnostic test, the average sensitivity and specificity would indicate that it would be of limited use. If undocumented impaired persons are assumed to be unimpaired, then specificity becomes excellent; however, in the light of mediocre sensitivity, this assumption seems to be questionable. Because there is considerable variation in documentation between Nordic hospital departments, there is reason to believe that this is of concern in acute care all over the world. The study confirms previous studies that have demonstrated deficient documentation of function and co-morbidities in medical records [5–9]. These deficits may adversely affect clinical care and research based on retrospective chart reviews [10]. Because of information overload and at times chaos in current hospital records, it might be helpful to have a systematic approach. Checklists have been shown to improve care [11]. Shortening the length of stay in hospitals requires early discharge planning [12]. If activities of daily living (ADL) and instrumental activities of daily living (IADL) capacities are not communicated and difficulties in these areas are not planned for during the hospitalisation, the patient is put at high risk of significant suffering and possible readmission [13]. Readmissions account for high cost and a possible waste of resources [14]. Between 21 and 65% of our patients had been hospitalised within 90 days of the index admission. Documentation is a precondition to recognition and action. For further discussion, see Appendix 5 in the supplementary data on the journal website (http://www.ageing.oxfordjournals.org/). We chose to use the MDS-AC documentation as a gold standard, because the instrument is reliable with excellent κ values for the majority of individual variables documented [4]. This we judged to be our best choice to present the data. Another discussion point is that the documentation of an issue is sought in the hospital record within 48 h of admission, whereas the MDS-AC evaluation was performed within 24 h. The study was concerned with co-morbidity and functional variables but not with the primary cause of admission; hence, we gave hospital staff more time to catch up. This approach does not negate the general message of this article. It is shown from the study results and discussion that the use of the MDS-AC instrument is one approach to improve data collection for various functional variables and co-morbidity that relate importantly to the outcome of care. A key question is whether the ‘cost’ in terms of staff time on a busy ward is justified in terms of the extra information that becomes available. There are several responses to this question. First, the MDS-AC instrument is under development and will be streamlined according to the experience of its use. Second, the future recommendation will be to make a full assessment only of those who fail the screening questions of a few variables. Third, currently, much information is not collected, but at the same time, clinical experience tells that there is an overlap in information collected by nurses and doctors. By avoiding double documentation with a clear division of tasks between physicians and nurses, time can be saved for more full documentation. Lastly, the future electronic medical record may call for us as professionals to put as much of our daily information into reliable form that then can be used for multiple purposes, such as care, quality assessment and research. The MDS-AC instrument is of that nature. Further studies are needed, preferably randomised and controlled, using patient outcomes as endpoints to prove the superiority of the MDS-AC approach to the traditional hospital record. This study has shown that during the first 48 h of acute care, co-morbidity or functional issues are not in the focus of care, and thus, those issues are not systematically documented in the medical record of older persons. These study results imply that the quality of documentation in acute care hospitals can be improved dramatically by relying on a standardised approach when it comes to co-morbidities and functional status. When a standardised, valid and reliable assessment instrument, the Minimum Data Set for Acute Care (MDS-AC), is used as a gold standard in comparison with traditional hospital record of older persons, it is shown that only about half of all information on co-morbidities and functional status is documented. These study results imply that the quality of documentation in acute care hospitals can be improved dramatically by relying on a standardised approach when it comes to co-morbidities and functional status. We thank the following for their financial support, from which all researchers are independent: The Nordic Lions Red Feather Fund; The Nordic Council of Ministers; The Roikjer Fund, Denmark; The Finnish Lions Fund, Finland; The Icelandic Lions Fund, Memorial Fund of Helgu Jónsdóttur and Sigurliða Kristjánssonar, The Reykjavik Hospital Research fund, The St. Joseph’s Research Fund, Iceland; The Norwegian Medical Society Quality Fund no. 2, Diakonhjemmet Hospital and Diakonhjemmet University College, The Diakonhjemmet Research Fund, Norway; and The Sweden’s Lions Fund, Sweden. All authors declare that the answers to the questions on competing interest are all no and therefore have nothing to declare. This study has been presented in abstract form at the 5th European Congress of Gerontology, 2–5 July 2003, Barcelona, Spain, and at the 17th Nordic Congress of Gerontology, 23–26 May 2004, Stockholm, Sweden. Nordic researchers who worked on the day-to-day collection of data were Eva Folkersen, Ghita Sell and Jytte Christensen of Denmark; Kaija Lindman, Päivi Putkonen, Tiina Heimola, Anja Teikari and Ulla Kuusi of Finland; S. Bjartmarz and O. Samuelsson of Iceland; Hege Aamilid of Norway and Ulla-Britt Johansson and Britt-Inger Norberg of Sweden. We specially thank Hrafn Pálsson, The Ministry of Health in Iceland, staff at all hospital sites participating in the study and InterRAI for use of the MDS-AC instrument (see http://www.interrai.org).
10
Comprehensive Discharge Planning and Home Follow-up of Hospitalized Elders
Mary D. Naylor, Dorothy Brooten, Roberta Campbell et al. · JAMA · 1999 · 1.8K citations
H. S. Lau, Christa Florax, A. J. Porsius et al. · British Journal of Clinical Pharmacology · 2000 · 258 citations · Full text