In total, 7212 Nova Scotians were identified as having died due to lung, colorectal, breast or prostate cancer over the six-year study period. Males comprised a larger proportion of deaths, as did adults aged 65 years and older, those who died due to lung cancer and survivors of at least 150 days from date of initial cancer diagnosis (Table ). The number of deaths across fiscal years remained relatively stable.
Characteristics of adults who died due to lung, colorectal, breast or prostate cancer in Nova Scotia between April 1, 1992 and March 31, 1998
In total, 139,641 visits or a median of 13 visits per patient (mean 19.4; standard deviation [SD] 20.3), were provided by FPs to patients during their end-of-life with 94% of patients receiving at least one FP visit. Variability across fiscal years was minimal, ranging from 92.9% receiving at least one FP visit in 1993–94 to 95.3% in 1997–98.
The majority of FP visits were provided to hospital inpatients (64%), followed by the office (15%), home (10%), the emergency department (5%) and long-term care (5%). Of visits provided to hospital inpatients, almost 71% were categorized as a 'subsequent hospital visit', which represents continuing in-hospital care. Other inpatient visits included initial hospital visits (4.7%), visits after four weeks (17.5%), supportive care visits (4.6%) and urgent or emergency care visits (2.5%).
Temporal trends associated with ambulatory and inpatient FP visits per 100 EOL person-days are shown in Figure along with the average number of days spent as hospital inpatient stay. Ambulatory visits by service location are illustrated in Figure . After accounting for age, sex and survival time, the total number of FP visits were found to have decreased significantly over the time period (p < 0.0001 declining from 15.3 visits per 100 EOL person-days in 1992–93 to 11.8 visits per 100 EOL person-days in 1996–97 followed by a small increase to 13.6 visits per 100 EOL person-days in 1997–98. This nonlinear trend is primarily due to the decline and then rise in the number of inpatient visits made by FPs over time. A closer examination of these inpatient visits by category (e.g., initial hospital visits, subsequent visits) did not reveal any major shift in the distribution of inpatient visit types over time. In contrast, total ambulatory visits remained relatively stable over the six-year time period with no evident significant time trends. Stratification of ambulatory visits by location of visit indicate a significant linear trend in emergency department visits over time (p < 0.01). After accounting for age and sex, patients in 1997–98 made 18% more emergency department visits than patients in 1992–93 (adjusted RR 1.18; 95% CI 1.05, 1.34). No association was evident across time for visits provided in the office, at home or within a long term care facility.
Family physician inpatient and ambulatory visits and length of hospital stay among advanced cancer patients over time
Ambulatory family physician visits to cancer patients by location over time
Examination of ambulatory FP visits provided during regular hours (8:01 am-5 pm) showed no significant change over time. However, ambulatory visits provided during 'after' hours (5:01 pm-8 am, weekends) were found to have increased significantly over the study period (p < 0.03). Compared to patients in 1992–93, patients in 1997–98 received 12% more ambulatory visits after hours (adjusted RR 1.12; 95% CI 1.01, 1.25). Although this increase represented a significant change over time, after hour visits were relatively few compared to the large number of other ambulatory visits and therefore exerted no impact on the overall temporal effect among all ambulatory visits.
Significant temporal trends were evident with respect to both the total number of hospital admissions experienced by the patient and the total number of days they spent as a hospital inpatient. A decline in hospital admissions was seen over time, from 1.2 admissions in 1992–93 to 1.1 admissions per 100 EOL person-days in 1997–98. Compared to 1992–93, patients in all subsequent years experienced fewer hospital admissions after accounting for sex and age. By 1997–98 patients experienced 13% fewer hospital admissions than patients in 1992–93 (adjusted RR 0.87; 95%CI 0.82, 0.93).
Over 85% of patients spent at least one day as a hospital inpatient. Patients spent on average a total of 22.7 days in hospital (SD 27.4; median 14 days; range 0–180 days) or 15.6 days per 100 EOL person-days. Total length of hospital inpatient stays declined from 18.6 days per 100 EOL person-days in 1992–93 to 14.8 in 1997–98. Results from the age and sex adjusted regression analysis indicate total length of hospital inpatient stays in 1997–98 were 21% shorter than experienced in 1992–93 (adjusted RR 0.79; 95%CI 0.71, 0.88).
In total, 82,575 visits were made to a medical specialty during the end of life. The number of visits ranged from one to 169, with a median of 11 visits per patient (mean 11.4; SD 13.5). Age and sex adjusted regression analysis indicate visits to a medical specialty did not change significantly over time.