2.0 Health System Performance

Printer-friendly version

2.0 Health System Performance

Accessibility

2.1 Wait Time for Hip Fracture Surgery

Definition
Proportion with surgery same or next day:
Risk-adjusted proportion of hip fracture patients aged 65 and older who underwent hip fracture surgery on the day of admission or the next day.
Proportion with surgery same, next day or day after:
Risk-adjusted proportion of hip fracture patients aged 65 and older who underwent hip fracture surgery on the day of admission, the next day or the day after that.

Method of Calculation
Denominator:
The number of hip fracture patients aged 65 and older who underwent hip fracture surgery in an acute care hospital.
Numerator:
Numerator is a subset of the denominator according to one of the two available definitions. Wait time is measured from the date of the first admission with hip fracture (index admission) to the date when hip surgery was received.

A Technical Note describes case selection and inclusion/exclusion criteria.
Two separate logistic regression models were fitted with age, sex and selected pre-admission co-morbid diagnoses (heart failure, ischemic heart disease, hypertension, COPD, diabetes with complications and cardiac dysrhythmia) as independent variables, one modeling the probability of having hip fracture surgery on the same/the next day and the second one modeling the probability of having hip fracture surgery on the same/next/day after. Coefficients derived from the logistic models are used to calculate the probability for each case. The expected number of patients in a region is the sum of these case probabilities for that region. The risk-adjusted proportion is calculated by dividing the observed number by the expected number of cases and multiplying by the Canadian average. A 95 percent confidence interval is also calculated and the method used to calculate confidence intervals is available upon request. Refer to the Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Operative delay in older patients with hip fracture is associated with a higher risk of postoperative complications and mortality. Wait time for surgery following hip fracture provides a measure of access to care. The wait time may be influenced by co-morbid conditions, hospital transfers and practice differences related to certain types of medications, like blood thinners. However, longer waits may indicate lack of resources, physician unavailability or/and other issues related to the access to care.

Standards/Benchmarks
Benchmark of hip fracture fixation within 48 hours was set by Federal, Provincial and Territorial governments in December 2005.

Data Source
Discharge Abstract Database (DAD), CIHI

Reference Period
April 1, 2008 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rate for Quebec is not available due to differences in data collection.

References
First Common Benchmarks Will Allow Canadians to Measure Progress in Reducing Wait Times
. Press release, December 12, 2005. http://www.health.gov.on.ca/english/media/news_releases/archives/nr_05/nr_121205.pdf

Health Canada, Final Report of The Federal Advisor on Wait Time, 2006.

Canadian Institute for Health Information, Waiting for Heath Care in Canada: What we know and what we don't know. Ottawa: CIHI, 2006

Weller I, Wai EK, Jaglal S, Kreder HJ. The effect of hospital type and surgical delay on mortality after surgery for hip fracture. J Bone Joint Surg Br 2005;87:361-366.

Bergeron E, Lavoie A, Moore L, Bamvita JM, Ratte S, Gravel C, Clas D. Is the delay to surgery for isolated hip fracture predictive of outcome in efficient systems? J Trauma 2006;60:753-757.

Canadian Institute for Health Information, Health Indicators 2007. Ottawa: CIHI, 2007.

Vidal EIO, Moreira-Filho DC, Coeli CM, Camargo KR, Fukushima FB, Blais R. Hip fracture in the elderly: does counting time from fracture to surgery or from hospital admission to surgery matter when studying in-hospital mortality? Osteoporosis International 2008; Oct 7.

Comments
The indicators are calculated in days rather than hours since the procedure time was not available in the DAD.

Note, due to differences in methodology this indicator may differ from similar indicators developed and reported by jurisdictions..

Appropriateness

2.2 Caesarean Section Rate

Definition
Proportion of women delivering babies in acute care hospitals by Caesarean section.

Method of Calculation
(Number of Caesarean sections/Number of deliveries (live births and stillbirths)) * 100

Denominator (Delivery):
Inclusion:

Delivery coded in any diagnosis field:

ICD-9
641-676 with a fifth digit of 1 or 2; 650 or V27

ICD-10-CA
O10 - O16, O21 - O29, O30 - O37, O40 - O46, O48, O60 - O69, O70 - O75, O85 - O89, O90 - O92, O95, O98, O99 with a sixth digit of 1 or 2; or Z37

Exclusion:
Delivery in which an abortive procedure* was recorded:

CCP
78.52, 86.3, 86.4, 87.0, 87.1 or 87.2

CCI
5.CA.88^^, 5.CA.89^^, or 5.CA.93^^

Numerator (Caesarean section):
The numerator is a subset of the denominator. Caesarean section is identified as any procedure code* of:

CCP
86.0-86.2, 86.8 or 86.9

CCI
5.MD.60^^

*Code may be recorded in any position. Procedures coded as cancelled, previous, out-of-hospital and "abandoned after onset" are excluded.

Interpretation

Caesarean section rates provide information on the frequency of surgical birth delivery relative to all modes of birth delivery. Since Caesarean section delivery increases maternal morbidity/mortality and is associated with higher costs, Caesarean section rates are often used to monitor clinical practices with an implicit assumption that lower rates indicate more appropriate, as well as more efficient care.

Standards/Benchmarks
Guidelines defining the appropriate indications for Caesarean section are available.

Data Source
Discharge Abstract Database (DAD), CIHI
Fichier des hospitalisations MED-ÉCHO, Ministère de la Santé et des Services sociaux.

Reference Period
April 1, 2008 - March 31, 2009.

Comprehensiveness
Available for all provinces and territories.

References
National Collaborating Centre for Women’s and Children’s Health. Caesarean Section. Clinical Guidelines April 2004.

Canadian Institute for Health Information, Giving Birth in Canada: Regional Trends From 2001-2002 to 2005-2006. Ottawa: CIHI, 2007.

Canadian Institute for Health Information, Giving birth in Canada. A regional profile. Ottawa: CIHI, 2004.

Joseph KS, Kramer MS, Marcoux S, Ohlsson A, Wen SW, Allen A, Platt R. Determinants of pre-term birth rates in Canada from 1981 through 1983 and from 1992 through 1994. New England Journal of Medicine 1998; 339:14341439.

Liu S, Rusen ID, Joseph KS, Liston RM, Kramer MS, Wen SW, Kinch R; Maternal Health Study Group of the Canadian Perinatal Surveillance System. Recent trends in caesarean delivery rates and indications for caesarean delivery in Canada. Journal of Obstetrics and Gynaecology Canada 2004; 26:735-742.

Shearer EL. Cesarean section: medical benefits and costs. Social Science & Medicine 1993; 37:1223-1231.

Liu S, Heaman M, Joseph KS, Liston RM, Huang l, Sauve R, Kramer MS. Risk of Maternal Postpartum Readmission Associated With Mode of Delivery. Obstetrics and Gynecology 2005; 105: 836-842.

Comments
Prior to 2001– 2002 rate, deliveries were based on adjusted newborn counts. Beginning with 2002 – 2003 rate, stillbirths are included in the delivery count. Prior to that, stillbirths were excluded.

Effectiveness

2.3 Ambulatory Care Sensitive Conditions (ACSC) Hospitalization Rate

Definition
Age-standardized acute care hospitalization rate for conditions where appropriate ambulatory care prevents or reduces the need for admission to hospital, per 100,000 population under age 75 years.

Method of Calculation
(Total number of acute care hospitalizations for ambulatory care sensitive conditions (ACSC) under age 75 years / Total mid-year population under age 75 years) * 100,000 (Age adjusted)

Inclusion criteria:
Based on a list of conditions developed by Billings et al, any one most responsible diagnosis code of:

  • Grand mal status and other epileptic convulsions
  • Chronic obstructive pulmonary diseases
  • Asthma
  • Heart failure and pulmonary edema
  • Hypertension
  • Angina
  • Diabetes

See Technical Note for codes used.

Exclusion criteria:
1. Individuals 75 years of age and older.
2. Death before discharge.

Interpretation
Hospitalization for an Ambulatory Care Sensitive Condition is considered to be a measure of access to appropriate primary health care. While not all admissions for these conditions are avoidable, it is assumed that appropriate ambulatory care could prevent the onset of this type of illness or condition, control an acute episodic illness or condition, or manage a chronic disease or condition. A disproportionately high rate is presumed to reflect problems in obtaining access to appropriate primary care.

Standards/Benchmarks
The “right” level of utilization is not known and large regional variations in the rate of hospitalization for these conditions exist.

Data Source
Discharge Abstract Database (DAD), CIHI;
Fichier des hospitalisations MED-ÉCHO, ministère de la Santé et des Services sociaux.

Reference Period
April 1, 2008 - March 31, 2009.

Comprehensiveness
Available for all provinces and territories.

References
Anderson GM. Common conditions considered sensitive to ambulatory care. In: Goel V, Williams JI, Anderson GM, Blackstein-Hirsch P, Fooks C, Naylor CD, (eds): Patterns of Health Care in Ontario, 2nd Ed., Canadian Medical Association, Ottawa, 1996:104-110.

Billings J, Anderson GM, Newman LS. Recent findings on preventable hospitalizations. Health Affairs 1996; 15:239-249.

Billings J, Zeital L, Lukomnik J, Carey TS, Blank AE, Newman L. Impact of socio-economic status on hospital use in New York City. Health Affairs 1993; 12:162-173.

Manitoba Centre for Health Policy and Evaluation (MCHPE). Ambulatory Care Sensitive (ACS) conditions.
http://www.umanitoba.ca/centres/mchp/concept/dict/ACS_conditions.htm

Comments
Beginning with 2006-2007 rate, the definition of the ambulatory care sensitive conditions indicator was refined to better align as a measure of primary health care. In the revised definition, diabetes component will only include diabetes with short-term complications or diabetes without mention of complication; angina, hypertension and heart failure components will exclude records where cardiac procedures were also coded. Rates based on the new definition were calculated for the previous years to allow for comparisons over time.

2.4 30-Day Acute Myocardial Infarction (AMI) In-Hospital Mortality Rate

Definition
The risk adjusted rate of all cause in-hospital death occurring within 30 days of first admission to an acute care hospital with a diagnosis of Acute Myocardial Infarction (AMI).

Method of Calculation:
Numerator: Number of deaths from all causes occurring in-hospital within 30 days of admission for AMI
Denominator: Total number of AMI episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age, gender, and select preadmission comorbid diagnoses as independent variables. Coefficients derived from the logistic model are used to calculate the probability of in-hospital death following AMI for each case (episode). The expected number of in-hospital deaths in a region is the sum of the case probabilities of that region. The risk adjusted mortality rate (RAMR) is calculated by dividing the observed number of in-hospital deaths of each region by the expected number of in-hospital deaths of the region and multiplying by the Canadian average in-hospital death rate.  A 95 percent confidence interval for the RAMR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to the Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
A lower risk-adjusted mortality rate following AMI may be related to quality of care or other factors. It has been shown that the 30-day in-hospital mortality rate is highly correlated (r=0.9) with total mortality (death in and out of hospital) following AMI (Tu et al., 1999). Inter-regional variations in 30-day in-hospital mortality rates may be due to jurisdictional and institutional differences in standards of care, as well as other factors that were not included in the adjustment.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Source
Discharge Abstract Database (DAD), CIHI

Reference Period
Rates are based on the 3 years of pooled data: April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Tu JV et al. Acute myocardial infarction outcomes in Ontario. In Naylor CD, Slaughter PM (eds). Cardiovascular Health & Services in Ontario: An ICES Atlas. Toronto: Institute for Clinical Evaluative Sciences. 1999; 84-100.

Tu JV et al. Acute myocardial infarction outcomes in Ontario (Methods Appendix). In Naylor CD, Slaughter PM (eds). Cardiovascular Health & Services in Ontario: An ICES Atlas (Technical and methods appendices). Toronto: Institute for Clinical Evaluative Sciences. 1999.

Comments
Beginning with the rates based on 2003-2004 to 2005-2006 data, AMI case selection criteria were revised to account for the fact that an increasing number of AMI patients are undergoing revascularization procedure (percutaneous coronary intervention or coronary artery bypass) at their index admission. In the case of revascularization procedure, AMI may not be coded as the most responsible diagnosis and these cases were previously excluded from the indicator. In addition, exclusion criteria were revised and patients with a length of stay of less than 3 days and discharged alive are no longer excluded. Comparison of rates for this time period with those of previous years should be made with caution.

These rates should be interpreted with caution due to potential differences in the coding of comorbid conditions across provinces and territories. 

2.5 30-Day Stroke In-Hospital Mortality Rate

Definition
The risk adjusted rate of all cause in-hospital death occurring within 30 days of first admission to an acute care hospital with a diagnosis of stroke.

Method of Calculation:
Numerator: Number of deaths from all causes occurring in-hospital within 30 days of admission for stroke
Denominator: Total number of stroke episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age, gender, type of stroke and select preadmission comorbid diagnoses as independent variables. Coefficients derived from the logistic model are used to calculate the probability of in-hospital death following stroke for each case (episode). The expected number of in-hospital deaths of a region is the sum of these case probabilities in that region. The risk-adjusted mortality rate (RAMR) is calculated by dividing the observed number of in-hospital deaths of each region by the expected number of in-hospital deaths of the region and multiplying by the Canadian average in-hospital death rate.  A 95 percent confidence interval for the RAMR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Stroke is a leading cause of death and long-term disability. Adjusted mortality rates following stroke may reflect the underlying effectiveness of treatment and quality of care. Inter-regional variations in the stroke mortality rates may be due to jurisdictional and institutional differences in standards of care, as well as other factors that are not included in the adjustment.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Sources
Discharge Abstract Database (DAD), CIHI.

Reference Period
Rates are based on the 3 years of pooled data: April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Mayo NE, Goldberg MS, Levy AR, Danys I, Korner-Bitensky N. Changing rates of stroke in the province of Quebec, Canada: 1981-1988. Stroke 1991; 22:590-595.

Mayo NE, Neville D, Kirkland S, Ostbye T, Mustard CA, Reeder B, et al. Hospitalization and case-fatality rates for stroke in Canada from 1982 through 1991: the Canadian collaborative study group of stroke hospitalizations. Stroke 1996; 27:1215-20.

Weir N, Dennis MS. Towards a national system for monitoring the quality of hospital-based stroke services. Stroke 2001; 32:1415-21.

Comments
Beginning with rates based on 2003-2004 to 2005-2006 data, case selection criteria for stroke were revised to include patients transferred to rehabilitation during their index admission. In this case, stroke may not be coded as most responsible diagnosis and these cases were previously excluded from the indicator. In addition, stroke resulting from occlusion of precerebral arteries is now included in the indicator. These cases were previously excluded since their identification was not possible in the ICD-9 coding system. Comparison of rates for this time period with those of previous years should be made with caution.

This indicator is based on the methodology used to calculate the 30-day Acute Myocardial Infarction in-hospital mortality rate. Rates should be interpreted with caution due to potential differences in the coding of comorbid conditions across provinces and territories.

2.6 Acute Myocardial Infarction (AMI) Readmission Rate

Definition
Risk adjusted rate of unplanned readmission following discharge for Acute Myocardial Infarction (AMI).  A case is counted as a readmission if it is for a relevant diagnosis and occurs within 28 days after the index AMI episode of care. An episode of care refers to all contiguous in-patient hospitalizations and same-day surgery visits.

Relevant diagnoses for assigning readmission cases:

  • Acute myocardial infarction 
  • Other acute and subacute forms of ischemic heart disease
  • Old myocardial infarction
  • Angina pectoris
  • Other forms of chronic ischemic heart disease
  • Conduction disorders
  • Cardiac dysrhythmias
  • Functional disturbances following cardiac surgery
  • Pneumococcal pneumonia
  • Other bacterial pneumonia
  • Bronchopneumonia, organism unspecified
  • Pneumonia, organism unspecified
  • Urinary tract infection

Method of Calculation
Numerator:
Number of AMI episodes with a readmission for a given year
Denominator:
Total number of AMI episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age, gender and multiple previous AMI admissions (2 and more) as independent variables. Coefficients derived from the logistic model are used to calculate the probability of readmission for each case (episode). The expected number of readmissions in a region is the sum of the case probabilities of that region. The risk adjusted readmission rate (RARR) is calculated by dividing the observed number of readmissions of each region by the expected number of readmissions of the region and multiplying by the Canadian average readmission rate. A 95 percent confidence interval for the RARR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to the Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Readmission rates provide one measure of quality of care. The risk of readmission following an AMI may be related to the type of drugs prescribed at discharge, patient compliance with post-discharge therapy, the quality of follow-up care in the community, or the availability of appropriate diagnostic or therapeutic technologies during the initial hospital stay. Although readmission for medical conditions may involve factors outside the direct control of the hospital, high rates of readmission act as a signal to hospitals to look more carefully at their practices, including the risk of discharging patients too early and the relationship with community physicians and community-based care.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Sources
Discharge Abstract Database (DAD), CIHI;
National Ambulatory Care Reporting System (NACRS), CIHI;
Alberta Ambulatory Care Database, Alberta Health and Wellness.

Reference Period
Rates are based on the 3 years of pooled data: April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Brown AD, Anderson GM. Methods for measuring clinical utilization and outcomes. In Baker GR, Anderson GM, Brown AD et al (eds.) The Hospital Report 99.  Health Care Performance Measurement Group, University of Toronto, Toronto, 1999.

Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Hospital Report Acute Care 2001. Technical notes, Clinical Utilization and Outcomes. Canadian Institute for Health Information and the University of Toronto.  A joint initiative of the Ontario Hospital Association and the Government of Ontario, 2001.

Comments
Beginning with rates based on 2003-2004 to 2005-2006 data, AMI case selection criteria were revised to account for the fact that increasing number of AMI patients are undergoing revascularization procedures (percutaneous coronary intervention or coronary artery bypass) at their index admissions. In the case of revascularization procedure, AMI diagnosis may not be coded as most responsible and these cases were previously excluded from the indicator. In addition, exclusion criteria were revised and patients with length of stay less than 3 days and discharged alive are no longer excluded. Comparison of rates for this time period with those of previous years should be made with caution.

All jurisdictions for which this indicator can be calculated are now coding to the ICD-10-CA/CCI standard. Therefore, an adjustment that was made to this indicator to enable comparison of rates between ICD-9/ICD-9-CM and ICD-10-CA jurisdictions for data years 2000-2001 to 2002-2003 and 2001-2002 to 2003-2004 has been discontinued. This adjustment was to exclude AMI cases that occurred in the 4-8 week period following an earlier AMI (ICD-9/ICD-9-CM labels these as acute AMI while ICD-10-CA considers such cases to be chronic). Effective with the rates based on 2002-2003 to 2005-2006, this exclusion was no longer applied and only AMI occurring within four weeks of a previous AMI is considered acute. Comparisons with the rates for the previous years should be made with caution.

A new “combination” code for acute lower respiratory infections in patients with Chronic Obstructive Pulmonary Disease (J44.0) was introduced with ICD-10-CA.  According to the Canadian Coding Standards, if COPD patients presented with pneumonia, only J44.0 should be used, but not the other codes from the J44 rubric. This code should be assigned as most responsible diagnosis (MRDx) with pneumonia assigned as a secondary diagnosis.  To correct the evident erroneous applications of this coding standard, pneumonia cases coded as MRDx will be removed if J44 was also recorded in any of the secondary diagnosis positions.

Effective with the rates based on the 2001-2002 to 2003-2004 data, the methodology for this indicator no longer excludes readmissions associated with a transfer for catheterization, angiography, angioplasty, insertion of pacemaker or coronary artery bypass graft surgery.  This change may affect the comparability of rates with those appearing in previous releases.

2.7 Asthma Readmission Rate

Definition
Risk adjusted rate of unplanned readmission following discharge for asthma.  A case is counted as a readmission if it is for a relevant diagnosis and occurs within 28 days after the index episode of care. An episode of care refers to all contiguous in-patient hospitalizations and same-day surgery visits.

Relevant diagnoses for assigning readmission cases:

  • Pneumococcal pneumonia
  • Other bacterial pneumonia
  • Bronchopneumonia, organism unspecified
  • Pneumonia, organism unspecified
  • Asthma
  • Empyema
  • Pulmonary collapse
  • Respiratory arrest
  • Respiratory complications during or resulting from a procedure

Method of Calculation
Numerator:
Number of asthma episodes with a readmission for a given year
Denominator:
Total number of asthma episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age, gender and multiple previous admissions for asthma (2 and more) as independent variables. Coefficients derived from the logistic model are used to calculate the probability of readmission for each case (episode). The expected number of readmissions of a region is the sum of these case probabilities in that region. The risk adjusted readmission rate (RARR) is calculated by dividing the observed number of readmissions of each region by the expected number of readmissions of the region and multiplying by the Canadian average readmission rate. A 95 percent confidence interval for the RARR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to the Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Readmission rates provide one measure of quality of care. Although readmission for medical conditions may involve factors outside the direct control of the hospital, high rates of readmission act as a signal to hospitals to look more carefully at their practices, including the risk of discharging patients too early and the relationship with community physicians and community-based care.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Sources
Discharge Abstract Database (DAD), CIHI;
National Ambulatory Care Reporting System (NACRS), CIHI;
Alberta Ambulatory Care Database, Alberta Health and Wellness.

Reference Period
Rates are based on the 3 years of pooled data:  April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Brown AD, Anderson GM. Methods for measuring clinical utilization and outcomes. In Baker GR, Anderson GM, Brown AD et al (eds). The Hospital Report 99. Health Care Performance Measurement Group, University of Toronto, Toronto, 1999.

Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Hospital Report Acute Care 2001. Technical notes, Clinical Utilization and Outcomes. Canadian Institute for Health Information and the University of Toronto. A joint initiative of the Ontario Hospital Association and the Government of Ontario, 2001.

Comments
A new “combination” code for acute lower respiratory infections in patients with Chronic Obstructive Pulmonary Disease (J44.0) was introduced with ICD-10-CA.  According to the Canadian Coding Standards, if COPD patients presented with pneumonia, only J44.0 should be used, but not the other codes from the J44 rubric. This code should be assigned as most responsible diagnosis (MRDx) with pneumonia assigned as a secondary diagnosis.  To correct the evident erroneous applications of this coding standard, pneumonia cases coded as MRDx will be removed if J44 was also recorded in any of the secondary diagnosis positions.

2.8 Hysterectomy Readmission Rate

Definition
Risk adjusted rate of unplanned readmission following discharge for hysterectomy.  A case is counted as a readmission if it is for a relevant diagnosis and occurs within 7 or 28 days (depending on condition) after the index episode of care. An episode of care refers to all contiguous in-patient hospitalizations and same-day surgery visits.

Relevant diagnoses for assigning readmission cases:

  • Acute posthemorrhagic anemia  - 28 days
  • Paralytic ileus - 28 days
  • Cardiac complications during or resulting from a procedure - 28 days
  • Respiratory complications resulting from a procedure - 28 days
  • Postoperative infection - 28 days
  • Urinary tract infection, site not specified  - 7 days
  • Retention of urine  - 7 days

Method of Calculation
Numerator:
Number of hysterectomy episodes with a readmission for a given year
Denominator:
Total number of hysterectomy episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age as an independent variable. Coefficients derived from the logistic model are used to calculate the probability of readmission for each case (episode). The expected number of readmissions of a region is the sum of these case probabilities for that region. The risk adjusted readmission rate (RARR) is calculated by dividing the observed number of readmissions of each region by the expected number of readmissions of the region and multiplying by the Canadian average readmission rate. A 95 percent confidence interval for the RARR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Readmission rates provide one measure of quality of care. Although readmission following surgery may involve factors outside the direct control of the hospital, high rates of readmission act as a signal to hospitals to look more carefully at their practices, including the risk of discharging patients too early and the relationship with community physicians and community-based care.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Sources
Discharge Abstract Database (DAD), CIHI;
National Ambulatory Care Reporting System (NACRS), CIHI;
Alberta Ambulatory Care Database, Alberta Health and Wellness.

Reference Period
Rates are based on the 3 years of pooled data:  April 1, 2006– March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Benbassat J, Taragin M. Hospital readmissions as a measure of quality of health care. Advantages and limitations. Archives of Internal Medicine 2000; 160:1074-1081.

Brown AD, Anderson GM. Methods for measuring clinical utilization and outcomes. In Baker GR, Anderson GM, Brown AD et al (eds).  The Hospital Report 99.  Health Care Performance Measurement Group, University of Toronto, Toronto, 1999.

Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Hospital Report Acute Care 2001. Technical notes, Clinical Utilization and Outcomes. Canadian Institute for Health Information and the University of Toronto.  A joint initiative of the Ontario Hospital Association and the Government of Ontario, 2001.

Makinen J, Johansson J, Tomés C, Tomés E, Heinonen PK, Laatikainen T, Kauko M, Heikkinen AM, Sjoüberg J. Morbidity of 10,110 hysterectomies by type of approach. Human Reproduction 2001; 16:1473-1478.

Measuring the quality of Pennsylvanias HMOs. A managed care performance report. Fiscal Year 1999. Technical Report. The Pennsylvania Health Care Cost Containment Council. July 2000.

Comments
Effective with the rates based on the 2006-2007 to 2008-2009 data, hysterectomy cases include both total and sub-total hysterectomies. Sub-total hysterectomy was not uniquely identified in the Canadian Classification of Health Interventions (CCI) versions 2001 and 2003; therefore rates based on 2001-2002 to 2005-2006 data included only total hysterectomies. Identification of sub-total hysterectomies became possible with version 2006 of CCI. For jurisdictions with higher volumes of sub-total hysterectomies comparability with the previous years might be affected.

2.9 Prostatectomy Readmission Rate

Definition
Risk adjusted rate of unplanned readmission following discharge for prostatectomy.  A case is counted as a readmission if it is for a relevant diagnosis or procedure and occurs within 28 days after the index episode of care. An episode of care refers to all contiguous in-patient hospitalizations and same-day surgery visits.

Relevant procedures for assigning readmission cases:

  • Operations on the ureter
  • Operations on the urinary bladder
  • Operations on the urethra
  • Other operations on the urinary tract
  • Operations on the prostate and seminal vesicles

Relevant diagnoses for assigning readmission cases:

  • Intestinal infections, other specified bacteria
  • Pneumonia, organism unspecified
  • Urinary tract infection, site not specified
  • Hematuria
  • Prostatic hypertrophy
  • Retention of urine
  • Pneumococcal pneumonia
  • Other bacterial pneumonia
  • Bronchopneumonia, organism unspecified
  • Cardiac complications during or resulting from a procedure
  • Respiratory complications resulting from a procedure
  • Postoperative infection

Method of Calculation
Numerator:
Number of prostatectomy episodes with a readmission for a given year
Denominator:
Total number of prostatectomy episodes in an 11-month period

A Technical Note describes the episode building and case selection.

A logistic regression model is fitted with age and select preadmission comorbid diagnosis as independent variables. Coefficients derived from the logistic model are used to calculate the probability of readmission for each case (episode). The expected number of readmissions of a region is the sum of these case probabilities for that region. The risk adjusted readmission rate (RARR) is calculated by dividing the observed number of readmissions of each region by the expected number of readmissions of the region and multiplying by the Canadian average readmission rate. A 95 percent confidence interval for the RARR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to Model Specifications for a list of variables entered in the model and coefficient values.

Interpretation
Readmission rates provide one measure of quality of care. Although readmission following surgery may involve factors outside the direct control of the hospital, high rates of readmission act as a signal to hospitals to look more carefully at their practices, including the risk of discharging patients too early and the relationship with community physicians and community-based care.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Sources
Discharge Abstract Database (DAD), CIHI;
National Ambulatory Care Reporting System (NACRS), CIHI;
Alberta Ambulatory Care Database, Alberta Health and Wellness.

Reference Period
Rates are based on the 3 years of pooled data:  April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
Brown AD, Anderson GM. Methods for measuring clinical utilization and outcomes. In Baker GR, Anderson GM, Brown AD et al (eds).  The Hospital Report 99.  Health Care Performance Measurement Group, University of Toronto, Toronto, 1999.

Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Statistics in Medicine 1995; 14:2161-2172.

Hospital Report Acute Care 2001. Technical notes, Clinical Utilization and Outcomes. Canadian Institute for Health Information and the University of Toronto.  A joint initiative of the Ontario Hospital Association and the Government of Ontario, 2001.

Lu-Yao GL, Albertsen P, Warren J, Yao SL. Effect of age and surgical approach on complications and short term mortality after radical prostatectomy- A population based study. Urology 1999; 54: 301-307.

Comments
A new “combination” code for acute lower respiratory infections in patients with Chronic Obstructive Pulmonary Disease (J44.0) was introduced with ICD-10-CA.  According to the Canadian Coding Standards, if COPD patients presented with pneumonia, only J44.0 should be used, but not the other codes from the J44 rubric. This code should be assigned as most responsible diagnosis (MRDx) with pneumonia assigned as a secondary diagnosis.  To correct the evident erroneous applications of this coding standard, pneumonia cases coded as MRDx will be removed if J44 was also recorded in any of the secondary diagnosis positions.

These rates should be interpreted with caution due to potential differences in the coding of comorbid conditions across provinces and territories.

Safety

2.10 Hospitalized Hip Fracture Event Rate

Definition
Age-standardized rate of new hip fractures admitted to an acute care hospital per 100,000 population age 65 and older. New event is defined as a first-ever hospitalization for hip fracture or a subsequent hip fracture occurring more than 28 days after the admission for the previous event in the reference period.

Method of Calculation
(Total number of new hip fracture events for persons age 65 and older) / Total mid-year population age 65 and older) * 100,000 (Age adjusted)

Numerator Inclusion Criteria:

  1. Hip fracture present on admission
    ICD-10-CA: S72.0, S72.1, S72.2; ICD-9/ICD-9-CM: 820.0-820.3, 820.8, 820.9 coded as diagnosis type (1),  or
  2. Age at admission 65 years and older
  3. Gender recorded as male or female
  4. Admission to an acute care institution
  5. Canadian resident

Numerator Exclusion Criteria:

  1. Records with an invalid Health Card Number or date of birth
  2. Records with an invalid admission date
  3. Hip fracture admissions within 28 days after the admission date of the previous hip fracture hospitalization
  4. Transfers1

Interpretation
Hip fractures represent a significant health burden for seniors and for the health system. As well as causing disability or death, hip fracture may have a major effect on independence and quality of life. Measuring occurrence of hip fractures in the population is important for planning and evaluating preventive strategies, allocating health resources and estimating costs.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Source
Discharge Abstract Database (DAD), CIHI;
Fichier des hospitalisations MED-ÉCHO, ministère de la Santé et des Services sociaux.

Reference Period
April 1, 2008 - March 31, 2009. 

Comprehensiveness
Available for all provinces and territories.

Comments
This indicator includes all new hospitalized hip fractures in the reference period encompassing first-ever and recurrent events. A person may have more than one hip fracture event in the reference period. Hip fractures not admitted to an acute care hospital and in-hospital hip fractures are not included in this indicator. Hip fractures occurring in a hospital are reported separately in the In-hospital Hip Fracture indicator.

References
Chevalley T, Guilley E, Herrmann FR, Hoffmeyer P, Rapin CH, Rizzoli R. Incidence of hip fracture over a 10-year period (1991-2000): Reversal of a secular trend. Bone 2007; 40:1284-1289.

Marks R, Allegrante JP, Ronald MacKenzie C, Lane JM. Hip fractures among the elderly: causes, consequences and control. Ageing Research Reviews 2003; 2:57-93.

1 If a subsequent hip fracture admission occurs on the same day as or prior to the discharge date of a previous hip fracture admission, it is considered as a transfer.

2.11 In-Hospital Hip Fracture Rate

Definition
Risk-adjusted rate of in-hospital hip fracture among acute care inpatients age 65 and older, per 1,000 discharges.

Method of Calculation
Numerator:
  Total number of discharges coded with in-hospital hip fracture for patients age 65 and older
Denominator:
Total number of discharges among inpatients age 65 and older

A Technical Note describes the selection of cases.

A logistic regression model is fitted with age, sex, whether a surgical procedure was provided, and the following preadmission comorbid conditions: cancer, seizure, stroke, delirium and other psychosis, trauma as independent variables. Coefficients derived from the logistic model are used to calculate the probability of in-hospital hip fracture for each case (episode). The expected number of in-hospital hip fractures of a region is the sum of these case probabilities for that region. The risk adjusted in-hospital hip fracture rate (RAR) is calculated by dividing the observed number of in-hospital hip fractures of each region by the expected number of in-hospital hip fractures of the region and multiplying by the Canadian average rate. A 95 percent confidence interval for the RAR is also calculated and the method used to calculate confidence intervals is available upon request. Refer to Technical Note for a list of variables entered in the model and to Model Specifications for the coefficient values.

Interpretation
Proposed by the Agency for Healthcare Research and Quality (AHRQ) and based on the Complications Screening Program, this indicator represents a potentially preventable complication resulting from an inpatient stay in an acute care facility. Variation in the rates may be attributed to numerous factors, including hospital processes, environmental safety, and availability of nursing care. High rates may prompt investigation of potential quality of care deficiencies.

Standards/Benchmarks
Benchmarks have not been identified for this indicator.

Data Source
Discharge Abstract Database (DAD), CIHI.

Reference Period
Rates are based on 3 years of pooled data:  April 1, 2006 – March 31, 2009.

Comprehensiveness
Available for all provinces and territories, except Quebec. Rates for Quebec are not available due to differences in data collection.

References
AHRQ Quality Indicators - Guide to Patient Safety Indicators
. Rockville, MD: Agency for Healthcare Research and Quality, 2003. AHRQ Pub.03-R203.

Comments
Effective with the rates based on the 2004-2005 to 2006-2007 data, in-hospital hip fracture rate is reported by the jurisdiction where hospitalization has occurred rather than by the jurisdiction of patient residence.  With this change the indicator will better reflect the concept of patient safety in the hospitals. This change may affect the comparability with the rates for the previous years.

“Fracture of bone following insertion of orthopaedic implant, joint prosthesis, or bone plate” (ICD-10-CA code M96.6) and in-hospital hip fracture coded in conjunction with an external cause of injury code of “misadventure during surgical or medical care” have been excluded because these events do not reflect patient safety in the context with which it is currently understood. These exclusions are applicable as of the rate based on the 2001-2002 to 2003-2004 data.

These rates should be interpreted with caution due to potential differences in the coding of comorbid conditions across provinces and territories.