Comparison of retrospective interviews and prospective diaries to facilitate fall reports among people with stroke

SIR—Monitoring falls is an important aspect of stroke rehabilitation. Retrospective methods include face-to-face or telephone interviews, postal questionnaires and medical note reviews [1–4]. Prospective methods include post cards, medical records, diaries and calendars as well as surveillance systems [1, 5–7]. Although prospective methods are considered preferable [1, 8, 9], it remains difficult to ascertain the accuracy of reporting methods as both may lead to over- or under-reporting [8, 10–17] and many factors influence recall [1, 8, 18–20]. Recall of falls in the previous year has reasonable sensitivity (80–89%) and specificity (91–95%) [5, 11, 16, 18] and produced better results than recall over shorter periods [16, 18] but recent studies have advocated short recall periods and intensive prospective (weekly or monthly) follow-up over longer periods [1, 5]. Little is known about the accuracy of fall reports among people with stroke. The present study examined the agreement between two fall-reporting methods (retrospective interviews and prospective fall diaries) over a 12-month period. This study formed part of a larger project predicting fall risk among stroke patients [21]. Ethical approval was obtained from the Southampton and South West Hampshire Local Research Ethics Committee. Consecutively hospitalised patients were recruited if they were independently mobile prior to stroke, able to give consent: those who failed a cognitive function test [22] which might have affected fall recall, were excluded. Two researchers carried out assessments: the first carried out tests of balance, function, mood and attention and was kept blind to participants’ fall status; the second collected data concerning falls, the focus of this paper. Retrospective falls data were collected during an interview with participants and significant others at 12 months following discharge from hospital to the community, using an interview schedule [23]. Over the same time period, prospective falls data were self-completed in a diary: participants (and significant others) were asked to record falls as and when they occurred and were reminded to do so by regular telephone calls and letters. We defined a fall as an event that resulted in a person coming to rest unintentionally on the ground or other lower level, not as a result of a major intrinsic event or overwhelming hazard [24]: participants were asked to adhere to this definition when reporting falls for either method. Participants were classified as fallers if they experienced one or more falls and as repeat fallers if they experienced two or more. Agreement between the retrospective and prospective methods of collecting numbers of falls was examined using kappa statistics and Bland and Altman limits [25], which give a range of values in which the difference is expected to lie 95% of the time. Response and falling rates are presented with 95% confidence intervals calculated in CIA [26]. Of the 122 participants recruited to the main study, retrospective falls information during the 12-month period following hospital discharge was available for 112 (93%, confidence interval 85–95%). Of these, 62 (55%, confidence interval 46–64%) reported one or more falls, and 45 (40%, confidence interval 32–49%) reported repeat falling. Using the prospective diary method, data for 76 (62%, confidence interval 53–70%) cases were available. Missing prospective information was due to the diary being lost (n = 26), death (n = 4) and no reason was recorded for the remaining 16. There were no participants with prospective but not retrospective information on falls and hence 76 cases were included in the comparison of the retrospective and prospective methods. On average they were aged 70 years; had spent 11 (median 9) weeks in hospital; and two-thirds were male. Their characteristics with respect to age, gender, stroke type, length of hospital stay and retrospectively collected falls were similar to the entire group (see Supplementary data available in Age and Ageing online). On first observation, the retrospective method resulted in a higher mean number of falls than the prospective method (5.9 vs. 1.7) though the median number of falls (1) was the same from both methods. There was a slightly higher percentage reporting one or more falls (61 vs. 57%) than two or more falls (41 vs. 41%) from the retrospective compared with the prospective method. The high mean (5.9) number of falls from the retrospective method was due to two participants who reported 100 and 200 falls retrospectively but no falls prospectively. These participants were falling on a frequent basis in the same circumstances but did not record these events in their diaries. In Figure 1, the Bland and Altman plot of the retrospective minus prospective difference in number of falls against their average is shown after excluding the two outliers, and there is then little difference between the two methods (retrospective minus prospective mean difference = 0.23, SD = 2.46). The Bland and Altman limits of agreement were −4.7–5.2, indicating that after excluding outliers, the remainder tend to give numbers of falls that are within 5 of each other. Bland and Altman plot comparing fall reports from retrospective interviews and prospective diaries (excluding outliers). Excluding outliers, the total number of falls reported was 182 (Table 1): of these 54 falls were reported retrospectively but not prospectively, and 37 prospectively but not retrospectively. In terms of number of participants involved, 16 reported excess retrospective falls, and 18 excess prospective falls (the remaining 40 reported the same number by both methods). In terms of the falls classification, the number of discrepancies was lower with only 8 classified as a faller retrospectively but not prospectively and 5 in the opposite direction, thus the methods agreed for 83% of participants with κ = 0.6481 indicative of good agreement. There were nine participants with discrepancies in each direction with respect to the classification as a repeat faller, and kappa = 0.5097 is indicative of only moderate agreement. Cross-tabulation of falls and participants reporting falling by the retrospective vs. prospective methods aCalculated under the assumption that falls reported by a participant by both methods are the same events. Cross-tabulation of falls and participants reporting falling by the retrospective vs. prospective methods aCalculated under the assumption that falls reported by a participant by both methods are the same events. In this study, fall information was collected both retrospectively during interviews and prospectively using diaries by 76 people with stroke in the year following hospital discharge. When outliers were excluded, the Bland and Altman limits of agreement indicated that the two methods are unlikely to differ by more than five falls. Even though the total numbers of falls reported were similar, they comprise 91 falls reported by both methods, and a further 91 reported by one method only (54 during interviews and 37 in the diaries). Relative agreement between the two methods in the general population has been reported by some [15, 16], but others reported greater disparities [18]. One major difference between the two methods in our study was the percentage for which data were available: 92% from the interviews, but only 62% from the diaries. Leaving the diary with people for the 12-month follow-up period was a limitation of the study design, in spite of the telephone and postal reminders: collecting the diaries in person or asking people to send them in on a monthly basis might have avoided this problem. Another potential problem with the prospective method was that two people who reported frequent falls in the same or similar circumstances during the retrospective interview did not bother to fill in their diaries possibly because they considered these falls to be an almost ‘normal’ occurrence. This led to a vast discrepancy in the total number of falls reported but also highlighted that face-to-face prompting can produce a more complete picture of patterns of falling in a small percentage of people. Though expert consensus has been reached recommending daily recording, continued surveillance of documentation and collection of circumstances of falls at least once per month for research studies [13], this combination of methods is too labour and time intensive for clinical settings. Even collecting fall information using prospective diaries combined with follow-up phone calls as utilised in the present study is not recommended for clinical settings, being costly in terms of manpower and time. If the time and money is available to collect prospective fall information, we recommend that regular face-to-face follow-up interviews are included in addition, to identify those who experience falls (particularly those with frequent falls) but fail to record them in their diary. On the basis of the findings from this study, we can recommend that structured retrospective interviews will yield acceptable results for most cases. More participants yielded information on falls from the retrospective interview than from self completed diaries. Frequent repeat fallers reported falls during the interview but did not record all falls in the diary. Excluding outliers, a similar number of falls were reported using either method. There was an 83% agreement in the classification of fallers between the two methods. None declared. This work was supported by The Stroke Association (Grant number TSA08/01).