Ministry 2002/03 Annual Service Plan Report -- Government of British Columbia.
   

Performance ReportingContinued

Goal 1: High Quality Patient Centred Care
Objective 1: Health services are delivered according to best practices and standards.
Strategies:

• Work with health authorities and providers to implement and audit health service standards.

Underway

• Introduce strategies to improve the care of people with chronic health conditions.

Underway

Achievements:

  • Performance agreements were established to outline expectations for health authorities' performance. A framework is being developed for monitoring levels of care, such as acute care and mental health services within each health authority.
  • BC established patient registries for diabetes and congestive heart failure so that health professionals can be proactive in scheduling tests and providing information.
  • Targets for rates of compliance with diabetes testing standards were established. More protocols and standards are under development for congestive heart failure, asthma and other chronic health conditions.
  • Chronic disease management Web site was established, including a secure Web site for practitioners to help assess level of care provided against BC standards.
  • Planning for physician collaboratives was completed. These will facilitate the development and distribution of evidence-based practice guidelines.

Performance Measures

Rates of compliance with selected protocols and standards (Goal 1: PM#1)

This indicator measures the number of patients with chronic illness who receive services that meet the standards and guidelines set by health professionals.

The 2002/03 measurement is of the percentage of patients with diabetes receiving at least 2 blood glucose (HbA1c) tests during the year, as set in BC's diabetes guidelines. Diabetes is one of the most common chronic diseases, affecting about five per cent of Canadians, and its prevalence is expected to increase significantly due to an aging population and rising rates of obesity. Blood glucose tests help flag potential complications for persons with diabetes.

Results

2002/03 Target 2002/03 Actual Status
Consider conditions and set targets in collaboration with the British Columbia Medical Association (BCMA) Targets were established to measure the percentage of patients with diabetes receiving at least 2 blood glucose (HbA1c) tests during the year.

Established
  Target 02/03: 45%
  Target 03/04: 55%
  Target 04/05: 60%
  Target 05/06: 65%

Met target

Percentage of Patients with Diabetes Receiving at least 2 Blood Glucose (HbA1c) Tests During the Year.

Notes: The analysis was based on individuals who:.
a. received 2 or more physician services (on different dates) paid by MSP within 365 days coded 250 (diabetes mellitus) since 1991; or
b. had 1 or more hospital discharges coded 250 since 1991; or
c. received insulin, oral hypoglycaemics, or testing strips paid for in part by PharmaCare since 1991; and
d. received one or more MSP-paid services in the 2001/02 fiscal year.

Source: BC Diabetes Registry, Population Health Surveillance and Epidemiology Branch, MSP Economic Analysis Branch.

 

Goal 1: PM#2a, 2b, and 2c

  • PM#2a: 30-day in-patient mortality or death rates for acute myocardial infarction (heart attack) and stroke
  • PM#2b: 5-year survival rates for lung, prostate, breast and colorectal cancer; relative survival rates for heart attack (365 days after admission to hospital) and stroke (180 days after admission)
  • PM#2c: Hospital re-admission rates for heart attack, congestive heart failure, pneumonia, and gastro-intestinal hemorrhage

Results

2002/03 Target 2002/03 Actual Status
Set targets Performance measures not carried forward See explanatory note below

Data for these three performance indicators cannot be provided within the required time frame of the Budget Transparency and Accountability Act and the Balanced Budget and Ministerial Accountability Act. As these measures are national health indicators, they will continue to be reported publicly in the Performance Indicator Reporting Committee and Canadian Institute of Health Information indictor reports.

Source: AMI and Stroke: BC administrative databases; Discharge Abstract Database / Hospital Morbidity database (CIHI); POI database, Vital Statistics files, and life tables (Statistics Canada); ISQ.

 

Goal 1: High Quality Patient-Centred Care
Objective 2: Care is provided at the appropriate level in the appropriate setting.
Strategies:  
• Rationalize and redesign hospital care. Underway
• Reallocate resources and develop policy to support innovative community, home care and palliative care services as alternatives to institutional care. Underway
• Provide 5,000 new home and community care placements (beds). Underway
• Implement an integrated health information system to facilitate better patient care across programs. Underway

Achievements:

  • Regional health authorities are at various stages of consolidating acute care services and creating inter-linked systems of small community hospitals or treatment centres for basic emergency services, larger community hospitals and regional referral centres.
  • A new Residential Access Policy was introduced to ensure that admittance to long-term care facilities is based on need.
  • The government passed the Community Care and Assisted Living Act to modernize regulation of care facilities, enable more local decision making to ensure appropriate care options, and support development of assisted living units.
  • Across the health authorities, 3,500 assisted living units are being provided under the Independent Living BC program, and additional units are being made available through other partnerships.
  • Finally, a strategic visionary plan and an annual tactical plan have been completed for developing a provincial information system that will improve patient care. A key goal is to connect health professionals and facilities to allow development of an electronic health record for each patient. This will enable better decisions about the health care provided and ensure appropriate care at all levels.

 

Click here to return to the top of this page.

 

Performance Measures

Rates of admission for conditions that could be managed outside hospital (classified as may not require hospitalization) (Goal 1: PM#2d)

May Not Require Hospitalization (MNRH) shows the rate of hospital admissions for conditions that could be managed outside hospital. This measure can help identify opportunities to reduce some categories of inpatient admissions and redirect care to others requiring hospital services.

2002/03 data from the Canadian Institute of Health Information (CIHI) is not available at this time due to delays in the cross-Canada implementation of a new classification scheme (ICD10-CA).

Results

2002/03 Target 2002/03 Actual Status
5% decrease
Baseline: 2000/01
Data for 2001/02 and 2002/03 not yet available Pending*
* Data for this measure will be available, pending the review and redevelopment of case mix groups by CIHI.

Source: Discharge Abstract Database, Information Support, Regional Programs, BC Ministry of Health; Population tables from PEOPLE26, BC Stats.

 

Percentage of home and community care clients with high care needs living in their own home (intermediate care level 3 or above) (Goal 1: PM#3a)

This indicator tracks the number of seniors and people with disabilities who have high care needs and receive home support or adult day care, allowing them to remain more independent and autonomous in their own home. Clients with high care needs are those who require intermediate or extended care — intermediate care level 3 or higher.

Results

2002/03 Target 2002/03 Actual Status
2% increase in the number of high care needs clients
Baseline: 2001/02
12.5% increase Exceeded target

Incremental Per Cent Change in the Number of High Care Need Clients Living in their Own Home.

  March 1999 March 2000 March 2001 March 2002 March 2003
Number of High Care Need Clients Living in their Own Home 6,292 6,414 6,523 7,370 8,292
Notes:
1.  Home and community care clients living in their own home refer to those receiving home support and adult day care.
2.  Those in residential care facilities refer to those in residential facilities, group homes and family care, and excludes mental health clients.
3.  High care needs clients are those clients at intermediate care level 3 or above.
4.  Counts of clients are made at a point in time. Data shown in the table are for March of each year.

Source: Continuing Care Information Management Systems, Continuing Care Data Warehouse.

 

Alternate Level of Care (ALC) days as a percentage of total inpatient days (Goal 1: PM#3b)

This measure indicates if patients have timely access to appropriate care in the most appropriate setting. The days that patients spend in hospital after their need for acute care has ended are called Alternate Level of Care (ALC) days. Patients remain in hospital longer than necessary for various reasons, including no available room in residential facilities or a delay in discharge arrangements. A reduction in ALC days results in more acute care beds being available for those who need acute care.

Results

2002/03 Target 2002/03 Preliminary* Status*
5% decrease
Baseline: 2001/02
6.1% decrease Exceeded target
Data are based on preliminary results to September 2002. Final results will be released in December in a supplement to this annual report.

Incremental Per Cent Change in the Proportion of Hospital In-patient Days that are Alternate Level of Care (ALC) Days.

  1999/2000 2000/2001 2001/2002 2002/2003* Target: 2002/2003
ALC Days 379,930 424,922 387,511 362,117  
Inpatient Days 2,750,961 2,730,992 2,613,268 2,608,313  
Per Cent 13.8% 15.6% 14.8% 13.9% 13.2%
Notes:
1.  2002/03 data are based on preliminary data to September 2002.
2.  Total inpatient days include acute care, rehabilitation and alternate level of care/discharge planning (no newborns).
3.  ALC days include all days coded alternate level of care or discharge planning.
4.  The data includes primary diagnosis and primary procedure only.

Source: Discharge Abstract Database, Information Support, Ministry of Health Services.

 

Appropriate use of blood products for clinical purposes measured by utilization of Intravenous/Immune Globulin (IVIG) Blood Products (Goal 1: PM#5)

This indicator measures the use of intravenous immune globulin (IVIG), a blood product made from human plasma. This blood product is licensed for treating some clinical conditions. However, it is also being used for treating conditions not supported by medical research. The average cost of IVIG therapy is approximately $70 per gram and the cost per course of treatment regularly exceeds $10,000. Through the Provincial Blood Coordinating Office, the ministry has implemented an IVIG Utilization Management Project to monitor the use of IVIG with the goal of stabilizing and reducing use of this treatment option.

Results

2002/03 Target 2002/03 Actual Status
Stable (at 66.9 grams per capita)
Baseline: 2001/02
Stable Met target

Utilization Rates for Intravenous/Immune Globulin (IVIG) Blood Products.

Source: Canadian Blood Services

 

Click here to return to the top of this page.

 

Administrative and support services expenditures as a percentage of total expenditures, by health authority (Goal 1: PM#8)

This indicator measures the amount that health authorities spend on administrative and support services, compared to their total expenditures. Administrative services include finance services, human resources and communications. Support services include maintenance, housekeeping, food services and security. This indicator helps health authorities plan their spending and ensure every available dollar is directed towards delivering patient care.

Results

2002/03 Target 2002/03 Preliminary* Status*
Maintain 2000/01 percentage (24.9%)
Baseline: 2000/01
Percentage decreased to 21.7% Exceeded target
Result is based on information available at the time of this report. Final results will be released in December in a supplement to this annual report.

Health Authority Administrative and Support Expenditures as a Percentage of Total Expenditures.

Notes:
1.  Result is based on information available at the time of this report. Final expenditure figures may be adjusted as health authorities finalize their fiscal year end accounts.
2.  The 2000/01 baseline was restated to 24.9% from 25.8%, which was published in the Ministry of Health Services' 2002/03 Service Plan. This change was a result of additional data being submitted to the ministry, plus changes resulting in consistency in the measurement of this indicator.
3.  Administrative and support services expenditures include functional centres or departments for administration and support as defined by CIHI, plus functional centres for nursing inpatient services administration, ambulatory care and community services administration, to ensure consistency with ministry analysis this excludes systems support.

Source: Health Authority Management Information System, Information Support Branch.

 

Goal 1: High Quality Patient-Centred Care
Objective 3: Patients have equitable and timely access to health care services.
Strategies:  

• Introduce strategies to improve access to basic health services (primary care).

Underway

• Modernize mental health care through the implementation of the Mental Health Plan.

Underway

• Commence implementation of a population needs-based funding formula to allocate resources to health authorities.

Completed

• Establish the Provincial Health Services Authority to reduce variability in access to specialized services across patient groups and place of residence.

Completed

• Work with the Ministry of Health Planning to implement the rural and remote health initiative.

Underway

• Expand the number of hospitals utilizing the PharmaNet system and BC BedLine.

Completed

Achievements:

  • The Primary Health Care Renewal Project, designed to provide better access to services, has finished the initial planning phase. Health authority plans will be implemented in 2003/04. These include investment in professional and organizational development, a range of practice models including multi-disciplinary practice teams, physician collaboratives and other shared care approaches.
  • A comprehensive mental health strategy was launched to: focus on early detection and evidence-based care; develop a permanent communications infrastructure to improve mental health literacy; better integrate mental health care throughout the health system; and focus on self-management and best practices.
  • The Provincial Health Services Authority (PHSA) was established to ensure specialized services are delivered effectively throughout BC. It is currently developing a Provincial Perinatal Service, which will optimize care of high risk pregnant women and their infants across the province. PHSA is also working with health authorities to examine patients' needs, provincial capacity and management of wait lists, under the Cardiac Program.
  • Programs such as the doctors' rural incentive agreement, rural specialist locum program, and expansion of medical school placements in regional universities encourage health professionals to live and work in rural BC.
  • All acute care hospitals in BC are currently using the BC BedLine call centre to request help in transferring patients. The BC BedLine Web site is also being used by hospitals that regularly receive patient transfers. Thirty hospitals are reporting their bed availability every four hours on the Web site, and another 11 are sending daily verbal reports.

Performance Measures

Improved continuity of care measured by the proportion of persons hospitalized for a mental health diagnosis who receive community or physician follow-up within 30 days of discharge (Goal 1: PM#4a)

This measures the percentage of persons aged 15–64 who have been hospitalized for a mental health illness and who receive at least one follow-up treatment at a community-based Mental Health Centre (MHC) or with a General Practitioner/Psychiatrist, within 30 days of being discharged from hospital. A high rate of community or physician follow-up indicates well-coordinated and accessible continuity of care for people with a mental health diagnosis.

Results

2002/03 Target 2002/03 Preliminary* Status*
3% increase
Baseline: 2000/01
3.1% increase Met target
Based on preliminary data to December 2002. Final results will be released in December in a supplement to this annual report.

Per Cent Change in Community Follow-up, within Thirty Days of Hospital Discharge.

Note:  2002/03 data are based on preliminary data to December 2002.

Source: Mental Health Data Warehouse and Discharge Abstracts Database, Information Support, BC Ministry of Health Services; MSP Claims.

 

Improved availability of community services measured by percentage of days spent by mental health patients in hospitals after the need for hospital care ended (Goal 1: PM#4bi)

This measures the number of days that Alternate Level of Care (ALC) mental health patients spend in acute care hospitals, as a proportion of all in-patient hospital days. Note that this measure is the same as the previous ALC measure, but focuses on people who are hospitalized for a mental health diagnosis.

Results

2002/03 Target 2002/03 Preliminary* Status*
No change (maintain or decrease rate below rate in baseline year)
Baseline: 2001/02
Rate decreased Exceeded target
Based on preliminary data to December 2002. Final results will be released in December in a supplement to this annual report.

Alternate Level of Care (ALC) Days, as a Per Cent of Hospital In-patient Days for Patients Hospitalized for a Mental Health Diagnosis.


For patients with a mental health diagnosis: 1998/1999 1999/2000 2000/2001 2001/2002 2002/2003
ALC Days 6,720 8,858 9,467 13,896 5,040
Inpatient Days 226,042 227,818 231,847 232,198 135,544
ALC Rate Per Cent 3.0% 3.9% 4.1% 6.0% 3.7%
Notes:
1.  2002/03 data are based on preliminary data to December 2002.
2.  Total inpatient days include acute care, rehabilitation and alternate level of care/discharge planning (no newborns).
3.  For 1999/00 and later, ALC includes all days coded alternate level of care or discharge planning.
4.  The data includes primary diagnosis and primary procedure only

Source: Discharge Abstract Database, Information Support, Regional Programs, BC Ministry of Health

 

Percentage of mental health clients receiving services in their own region (Goal 1: PM#4bii)

This measures the range of services that mental health clients receive in their own communities or regional health authorities. It indicates better access and greater availability of mental health care and community services for clients.

Results

2002/03 Target 2002/03 Preliminary* Status*
Increase (above 85.8%)
Baseline: 2001/02
Increased Met target
Based on partial data to January 2003. Final results will be released in December in a supplement to this annual report.

Per Cent of Mental Health Clients Receiving Services in their Own Health Authority.

Note: 2002/03 data are based on preliminary data to January 2003.

Source: Mental Health Research Data Warehouse, Information Support, BC Ministry of Health Services.

 

Proportion of mental health clients accessing community services (Goal 1: PM#4biii)

This indicator shows the availability of community services by measuring the percentage of people with a mental disorder who have received at least one mental health service from a care provider or health centre in the community.

Over the past five years, an increasing proportion of clients with mental illnesses are receiving care through community-based services — at a mental health centre, a physician's office or both.

Results

2002/03 Target 2002/03 Actual Status
Increase
Baseline: 2000/01
Data not yet available Pending*
Final results or data for 2002/03 will be released in December in a supplement to this annual report.

Proportion of People with Mental Health Diagnoses Accessing Community Services.

Source: Mental Health Research Data Warehouse, Information Support, BC Ministry of Health Services.

 

Waiting times for key services — Radiotherapy and Chemotherapy (Goal 1: PM#6a)

Monitoring wait times helps ensure patients' cancers are treated as early as possible to achieve the best outcomes. This indicator measures the percentage of patients that begin radiotherapy within four weeks of being ready to treat and the percentage of patients who start chemotherapy within two weeks of being ready to treat. The measurement is based on median wait times for these services and the number of patients treated.

As of March 2003, 87 per cent of patients requiring radiotherapy started treatment within four weeks of being ready to treat, slightly below the target of 90 per cent. The 2002/03 target value was set based on inaccurate baseline data for 2001/02, which estimated radiotherapy treatment at 88 per cent in that year when it was actually 82 per cent. Although the 2002/03 treatment rate remains below the 90 per cent target, results show a significant and steady improvement over 2000/01, when only 72 per cent of patients began treatment within four weeks. For chemotherapy, consistently, the BC Cancer Agency reports that over 90 per cent of patients receive treatment within two weeks.

The expansion of the Victoria Cancer Clinic provides more radiotherapy services in BC to meet the increasing demand for these services. Planning for a new clinic in Abbotsford will help ensure provincial services keep up with the growing incidence of cancer.

Results

2002/03 Target 2002/03 Actual Status
Radiotherapy:
90% of patients begin treatment within 4 weeks of being ready to treat
Radiotherapy:
87% of patients began treatment within 4 weeks of being ready to treat
Missed target*
Chemotherapy:
90% of patients begin treatment within 2 weeks of being ready to treat
Chemotherapy:
Over 90% of patients began treatment within 2 weeks of being ready to treat
Met target
See discussion above for an explanation of the variance.

Per Cent of Patients Receiving Radiotherapy Treatment within 4 Weeks of Being Ready to Treat.

Source: BC Cancer Agency

 

Regional variation in access to selected services (Goal 1: PM#7)

Measuring regional rates for major surgical procedures helps compare access to services among health authorities (HA): the wider the variation, the greater the inequities in accessing services across health authorities. This indicator specifically measures utilization rates of hip, knee and cardiac surgeries for regions. Comparative data for 2002/03 showing variations across BC's regions will be available in December 2003.

Results

2002/03 Target 2002/03 Actual Status
Reduce variation in regional differences to access services by 10%
Baseline: 2001/02
Data for hip, knee and cardiac surgeries will not be available for 2002/03 until December 2003 Pending*
Final results will be released in December in a supplement to this annual report.

 

Click here to return to the top of this page.

 

  Hip Replacement
Surgeries
Knee Replacement Surgeries Coronary Artery Bypass Surgeries
  2001/02 HA Range 2002/03 Target Range 2001/02 HA Range 2002/03 Target Range 2001/02 HA Range 2002/03 Target Range
Range in Surgery Rates Across Health Authorities (cases per 1,000 population) 30.7 27.6 52.0 46.8 19.0 17.1
Notes:
1.  The surgery and procedure rates are per 100,000 age-adjusted population in the health authority.
2.  TResidents of BC treated out-of-province are included. BC non-residents are excluded.

Source: Discharge Abstract Database, Information Support, Ministry of Health Services.

 

24/7 access to basic health services (primary care) (Goal 1: PM#10)

24/7 access to services is vital to improving primary health care and building a patient-centred health system in BC. Usage rates for the BC NurseLine indicate access to skilled care providers — to specially trained registered nurses — by telephone, anytime of the day or night, across the province. Further, the practice among many physicians is to direct their patients to hospital emergency departments after hours. The NurseLine provides an alternative for obtaining an objective assessment of patients' problems.

Results

2002/03 Target 2002/03 Actual Status
25% increase in NurseLine use 43% increase in NurseLine use Exceeded target
15% increase in after hour call-forwarding from physician offices to NurseLine
Baseline: 2001/02
203% increase in after hour call-forwarding from physician offices to NurseLine Exceeded target

Number of Calls to the NurseLine Number of Times Physicians Call Forwarded After Hours to the NurseLine
2001/02       99,497 2001/02      335 (revised 01/02)        
2002/03     142,142 2002/03   1,015                              

Source: BC Ministry of Health Services

 

Goal 1: High Quality Patient-Centred Care  
Objective 4: Patient satisfaction and public confidence in the system are increased.  
Strategies:

• Implement an annual mental health report card.

Underway

• Increase access to information to help patients and their families understand and manage their health through the self-care project, the NurseLine and other patient self-care approaches.

Completed

Achievements:

  • BC HealthGuide Program: Since April 2001, over 1.8 million books have been distributed to households, health professionals, and health related agencies in BC; more than 1,000,000 hits have been made to BC HealthGuide OnLine; and 27,000 BC First Nations Health Handbooks were distributed to aboriginal communities throughout BC.
  • BC NurseLine: Since opening, the BC NurseLine has provided health information and advice to over 240,000 callers.
  • Through the Provincial Chronic Disease Strategy, the province gives the public and service providers Web-based access to information and tools to manage chronic diseases.

Performance Measures

Public satisfaction rates (Goal 1: PM#9)

This indicator measures peoples' level of satisfaction with health services in general and with services received in a hospital, doctor's office or community centre. This information helps identify improvements from the patient's perspective, and consequently supports system accountability and a shift towards a patient-centered health system.

Original targets were selected based on a 2001 Leger Survey1. This survey indicated that 56.6 per cent of Canadians were satisfied with the health system, compared with 41.1 per cent of respondents who were not satisfied. In BC, 46.6 per cent of respondents were satisfied with the health system. The Leger Survey has not been repeated.

The recently released 2003 Aventis Healthcare Survey, which surveyed 1,500 people with employer-sponsored health plans, found that 23 per cent of respondents said the health care system was excellent/very good and a further 58 per cent said it was good. Comparable figures at the national level were 30 per cent and 52 per cent respectively.


1 Leger Marketing surveyed 1,507 Canadians between June 5 and 13, 2001; the results are considered accurate within 2.6% points, 19 out of 20.

Results

2002/03 Target 2002/03 Actual Status
Maintain public satisfaction levels at current rate of 45 – 50% satisfied with the way health services are provided Comparable 2002 data not available
Similar data show: 23% rate health care system as excellent/good
58% rate health care system as good
Met target

Source: The 2003 Aventis Healthcare Survey; Research, Rogers Healthcare and Financial Services Group.

 

 
Home -- 2002/03 Annual Service Plan Reports.
Back.
Feedback. Privacy. Disclaimer. Copyright. Top. Government of British Columbia.