Surgical mortality review reduces preventable deaths and patient safety indicators (PSIs)

Abstract

Financial penalty for unacceptable rates of mortality and Patient Safety Indicators (PSIs) is the reality of external quality management. Review of hospital deaths have shown a decrease in mortality ratios, but there are little data evaluating the long-term viability and results of these programs. Methods: From 1/2013 through 8/2014, 26,699 inpatients were cared for on 12 surgical services. A surgeon from each service led reviews of all mortality and PSIs with central reporting of preventability. We compared the proportion of preventable mortality, PSIs and the UHC Observed to Expected Mortality ratio (O:E ratio) over time. Statistical significance was p<0.05 by Poisson regression. Results: Of the 26,699 inpatients in the study period, there were 510 deaths (1.9%) and 553 PSIs (2%) reviewed. Of the 510 deaths, 137 were categorized as possibly preventable or preventable. The odds ratio of a preventable mortality was half in the final quarter compared to the first quarter and this reduction was primarily seen in high-risk services (p<0.05). The proportion of preventable PSIs fell from 70% to 26% during the same time period (p<0.05). The O:E ratio was consistently below 1(less deaths than expected) and fell throughout the study period (p<0.05). Conclusions: The improvement in the O:E ratio previously seen with hospital mortality review is a sustainable process. With a long-term commitment, the additional benefits are a reduction in preventable mortality and PSIs, especially in high-risk services. This process is one component that improves outcomes and reduces patient mortality. Internal Medicine Review Surgical mortality review reduces preventable deaths and patient safety indicators (PSIs)August 2017 2

Implementation of Hospital Mortality Reviews: A Systematic Review

“…Reported rates of preventable deaths in this review showed high variability, which is comparable with preventable death rates reported in previous publications. 1,4,6,38 Substantially higher rates of preventability were found in 2 publications 18,31 including only surgical patients from the same hospital (27.2%-28.9%) compared with publications 17,25,28 including all inpatient deaths within a specific timeframe (0.1%-8.0%). This is also comparable with the review by Rodwin et al, 6 who found higher mortality rates in cohorts of selected patients (range, 0.5%-51.1%) than in publications, which included all inpatient deaths consecutively (2.2%-4.1%).…”

“…Number of deaths were reported by 16 publications [16][17][18][19][20][21][22][23][25][26][27][28][29][30][31][32] and ranged from 66 to 7,856 (median, 1068 deaths). Observed mortality was reported in 12 publications 15,16,18,[20][21][22][23][26][27][28]30,31 and ranged from 0.4% in one study for all consecutive hospital deaths within 10 years to 7.8% in another study for all under-five child deaths within 12 months (median, 2.9%). Six publications 15,18,19,27,29,31 reported a decrease in mortality throughout the study period, whereas 1 study 25 reported no change in mortality rates.…”

“…Ten publications 15,17,19,20,[24][25][26][27][28][29]32 included all inpatient deaths in a specific timeframe, while 5 publications 16,[21][22][23]30 focused on inpatient child deaths, and 2 publications 18,31 included surgical inpatient deaths only. Number of deaths were reported by 16 publications [16][17][18][19][20][21][22][23][25][26][27][28][29][30][31][32] and ranged from 66 to 7,856 (median, 1068 deaths). Observed mortality was reported in 12 publications 15,16,18,[20][21][22][23][26][27][28]30,31 and ranged from 0.4% in one study for all consecutive hospital deaths within 10 years to 7.8% in another study for all under-five child deaths within 12 months (median, 2.9%).…”

“…During forward and backward citation tracking of these publications, we identified another 5 potentially relevant publications. Of these, we included another 2 publications 31,32 from one new study and one study already found in the screening in this review, respectively. Finally, 15 studies reported in 18 publications [15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32] were included in this review, of which 4 publications 16,18,28,31 contained quantitative data only, and 14 publications 15,17,[19][20][21][22][23][24][25][26][27]29,30,32 included both quantitative and qualitative data and were thus considered mixed method studies.…”

“…Observed mortality was reported in 12 publications 15,16,18,[20][21][22][23][26][27][28]30,31 and ranged from 0.4% in one study for all consecutive hospital deaths within 10 years to 7.8% in another study for all under-five child deaths within 12 months (median, 2.9%). Six publications 15,18,19,27,29,31 reported a decrease in mortality throughout the study period, whereas 1 study 25 reported no change in mortality rates. Eleven publications 16,17,[20][21][22][23][24]26,28,30,32 did not report mortality trends.…”