Mortality and cause of death in a cohort of people who had ever injected drugs in Glasgow: 1982-2012.

BACKGROUND: To describe all-cause and cause-specific mortality in a cohort of people who had ever injected drugs (PWID) with a low prevalence of HIV over 20-30 years. METHODS: Using a retrospective study design, identifying data from a cohort of PWID recruited between 1982 and 1993 through in-patient drug treatment services were linked to National Records for Scotland deaths data using probabilistic record linkage. We report all-cause and cause-specific mortality rates; standardized mortality ratios (SMR) across time, gender and age were estimated. RESULTS: Among 456 PWID, 139 (30.5%) died over 9024 person-years (PY) of follow-up. Mortality within the cohort was almost nine times higher than the general population, and remained elevated across all age groups. The greatest excess mortality rate was in the youngest age group, who were 15-24 years of age (SMR 31.6, 95% CI 21.2-47.1). Drug-related deaths declined over time and mortality was significantly higher among HIV positive participants. Although SMRs declined with follow-up, the SMR of the oldest age group (45-60) was 4.5 (95% CI 3.0-6.9). There were no significant differences in all-cause mortality rates between participants who were 25 years and older at cohort entry compared to younger participants. CONCLUSION: Mortality rates remained higher than the general population across all age groups. Screening services that identify a history of injecting drug use may be an opportunity to address risk factors faced by an ageing population of PWID and potentially have implications for future health care planning.

Centralization of services and reduction of adverse events in pancreatic cancer surgery.

BACKGROUND: The perioperative period is critical in the outcome for patients with pancreatic cancer. The aim of the present analysis was to examine adverse events in patients dying under surgical care in relation to changes in the organization of pancreatic cancer surgery. METHODS: From 1996 to 2005, 1,033 patients with pancreatic cancer, mean age of 71 years (range 21-97 years) died under surgical care. The incidence, mortality, and number of operations for pancreatic cancer remained stable across the time period, but the proportion of patients undergoing surgery in the five specialist cancer centers increased from 50 to 80 % (p < 0.001). Prior to death 260 (25 %) patients underwent operation and 96 (9 %) had endoscopic retrograde cholangiopancreatography (ERCP). There was a significant rise in ERCP (p = 0.03) and a decrease in non-resectional operations (p = 0.001). RESULTS: Since 1996, 52 (15 %) patients in whom 90 adverse events were recorded died following surgical intervention: 28 adverse events related to the perioperative period with 15 due to direct procedure complications such as bleeding or anastomotic leak; 13 were attributed to decision making around the choice or timing of the procedure. The postoperative mortality after curative pancreatic resection reduced from 3.5 to 1.8 %. Identified adverse events fell significantly in patients who died relating to the operative period (median of 3 per annum [1994-2000] to 1 per annum [2001-2005]) (p = 0.014) and medical care (3-0) (p = 0.003). CONCLUSIONS: Continuous peer review audit has demonstrated a reduction in the number of adverse events in patients dying with pancreatic cancer under surgical care as increased numbers of patients treated in specialist cancer centers.