Additional questions may be added as the challenge proceeds. Team leads will be notified if additional questions are added.
NOTE: The guiding specifications describe the elements of an ideal analysis. Teams are encouraged to submit analyses using the data available to them.
To what extent does re-prescribing and re-dispensing of culprit drugs contribute to the burden of adverse drug events presenting to Canadian hospitals?
Adverse drug events are unintended and harmful events related to medication use. A prospective Canadian study published in 2008 found that 12% of visits to a Canadian tertiary care emergency department were the result of an adverse drug event. Of these, 36% required admission to hospital, and almost 70% were deemed preventable. Data from Australia indicate that among patients admitted to hospital for adverse drug events, 30% may constitute repeat events. A prospective Dutch study found that among patients admitted to hospital for adverse drug events, 27% of culprit drugs—those implicated in causing the adverse events—were re-prescribed within only 6 months. It is unknown to what extent re-prescribing and re-dispensing of culprit drugs contributes to the burden of adverse drug events presenting to Canadian hospitals. Canadian data is required to understand the rate of re-prescribing and re-dispensing of culprit drugs to design, evaluate and implement effective strategies for prevention.
The analysis will identify the proportion of culprit drugs that are re-prescribed or re-dispensed over a minimum observation period of 6 months. A culprit drug is defined as a drug, or drug of the same class, that has previously caused an adverse drug event. The index adverse drug event must have been severe enough to either cause an emergency department visit or to require hospitalization.
The outcome adverse drug event must be validated in the data set used, and sensitivity and specificity for detecting adverse drug events must be reported.
The number of re-prescribed or re-dispensed culprit drugs over the observation period. When numerous drugs are implicated in an adverse drug event, the re-prescription of the combination of drugs contributing to that event should not contribute to the numerator more than once, unless they are discrete events over time that can be taken into account in a time to event analysis.
The total number of adverse drug events that are severe enough to either cause an emergency department visit or require hospitalization.
Minimum sample size of 100 adverse drug events in the denominator. Adult patients.
The analysis should include an observation period of at least six months after the occurrence of the adverse drug event, and how often a culprit drug was re-prescribed or re-dispensed over the period of observation.
This question presented in collaboration with HealthCareCAN.
What is the current antimicrobial consumption in a particular sector/unit/common condition? Are there particular outliers if we compare patient consumption, and what should we expect to see?
Drug-resistant infections are increasing as microbes adapt to the overuse of antimicrobials, resulting in significant morbidity and mortality. It is estimated that as much as 50% of antimicrobial usage may be unnecessary. Promoting prudent antimicrobial use in Canadian hospitals and healthcare organizations is essential to addressing antibiotic resistance. In order to determine which sectors/units/common conditions are at greatest risk of antimicrobial resistance and where we can improve inappropriate use and prescribing, we must first have an understanding of the current state of antimicrobial consumption. Having data which can inform us of current state, and which can be used for benchmarking will further this objective.
What is the antimicrobial consumption as expressed in DDD (defined daily dose) per 1000 patient days for each type of patient sector/unit/common condition? Sectors could be defined as: ambulatory care, acute care, and complex continuing care and rehabilitation. Unit can be defined: intensive care, surgical care, internal medicine. Common conditions can be defined as UTI (urinary tract infection), URTI (upper respiratory tract infections), pneumonia, skin and soft tissue infection (e.g. cellulitis).
The ideal response would be to provide consumption DDD/1000 patient day for a sector/unit/ common condition.
Suicide is the second leading cause of death among youth in Canada, and mental illnesses such as depression are the main risk factors of suicide. Youth ages 18 to 24 have the highest rates of mental illness such as depression and generalized anxiety disorder than other age groups.
With response rates to traditional surveys in decline, it is incumbent on governments and their partners to tap into new and emerging sources of publicly available data such as social media. Response rates to traditional surveys among youth, in particular male youth, are the some of the lowest among all age groups in the country.
Youth are active uses of social media, with about 74% of Twitter users between the ages of 15 and 25 years of age. Use of social media outlets such as Facebook or Twitter as a data source would serve to augment existing survey and administrative data sources and allow for better analysis, contextualization and interpretation of traditional incidence, prevalence and case level data currently available. There is even a potential opportunity that these new sources could provide an early indication of possible trends to guide more formal surveillance activities.
To answer this question teams should provide a rate or a proportion.
To assess whether non-traditional data sources would be a good source to supplement estimates from traditional surveys teams should provide an assessment of the quality of the data they are using from this new source in terms of sample size, representativeness and reliability.
The identification of potential new data sources must meet the following requirements:
- The data source must be currently available;
- Data extraction needs to ensure safeguards to individual privacy and confidentiality;
- Sample size and time period must be sufficiently meaningful to make the analysis relevant;
- The data source must not be currently used by federal/provincial/territorial governments in a systematic way; and,
- The data must pertain to the Canadian population.
Millions of new prescriptions are written each year, but for a variety of reasons a substantial proportion are never dispensed, or not picked up by, or on behalf of, the patient. This is known as primary nonadherence (i.e. not filling a first prescription for a drug). Knowing the volume and type of prescriptions that fall in this category would inform medication management practice and policy.
The analysis will identify the proportion of new (first-fill) prescriptions that are dispensed and picked up by the patient within 30 and 90 days. Refills should be excluded.
The analysis should be reported in the following fashion:
• Total number of new prescriptions written
• Proportion of new prescriptions that make it to the pharmacy for dispensing
• Proportion of new prescriptions that are picked up by patients
Supplementary analysis by class of medications and/or by reason for the prescription (e.g. on-going management of a chronic condition) would add value.
The analysis should include the number of prescriptions written by community-based health care providers counted in the denominator that are not dispensed within 30 (90) days, as well as the number of prescriptions written by community-based health care providers counted in the denominator that are not picked up by, or on behalf of, the patient within 30 (90) days.
Total number of prescriptions written by community-based health care providers during a given time period.
Ideal sample size of 50,000 or more prescriptions. Note: veterinary prescriptions should be excluded.
The time frame covered by the denominator should be at least one year, to avoid any potential seasonal effects. The analysis should include an observation period of at least 90 days after the last prescription in the denominator was written.
What is the reduction in microbial use in hospitals with microbial stewardship programs?
Drug-resistant infections are increasing as microbes adapt to the routine use of antibiotics, resulting in significant morbidity and mortality. It is estimated that as much as 50% of antibiotic usage may be unnecessary. Promoting prudent antibiotic use in Canadian hospitals and healthcare organizations is essential to addressing antibiotic resistance.
This question seeks to understand how large a reduction in antimicrobial usage is achieved – on average – by antimicrobial stewardship programs in Canada, and whether there are there particular outliers, indicating programs that have fared particularly well or poorly.
The analysis should indicate what is the average percent reduction in daily doses of therapy (DDT) for antimicrobials in the years before and after stewardship programmes were adopted in Canadian hospitals. The benchmark for adoption of a stewardship program is compliance with Accreditation Canada’s required organizational practice for antimicrobial stewardship (http://accreditation.ca/sites/default/files/rop-handbook-2016-en.pdf – page 39).
The ideal response would be an average of percent reductions in antimicrobial use for a representative sample of Canadian hospitals that have undertaken antimicrobial stewardship programs over the past decade. It may be useful to note that Accreditation Canada recently added stewardship policies as required organizational practice in Canadian hospitals beginning in 2014.
What proportion of acute opioid users become chronic users? What is the rate of opioid-related deaths among chronic opioid users?
While prescription opioids, are an effective treatment for acute pain, the evidence for their use in non-cancer related chronic pain management is weak. The downsides of chronic opioid use are numerous and have been the subject of recent news headlines. Understanding the relationship between acute and chronic use of opioids for non-cancer related pain and opioid-related deaths in chronic users would help to identify the scope of the issue for policymakers.
- The analysis should determine the proportion of acute opioid users that become chronic opioid users. Opioid use for cancer-related pain should be excluded.
- Numerator: Number of individuals with at least 1 ongoing prescription for opioids at six months and at 12 months.
- Denominator: Number of individuals with 1 or more new prescriptions for opioids within a pre-defined timeframe.
- What is the rate of opioid-related deaths among chronic opioid users?
- Numerator: Number of opioid-related deaths among chronic opioid users within specified timeframe.
- Denominator: Number of individuals with at least 1 ongoing prescription for opioids at six months and at 12 months.
Minimum sample size of 10,000 individuals with 1 or more new prescriptions for opioids.
Is there a difference in outcomes for patients with diabetes and those who do not have diabetes using ACE or ARBs?
Angiotensin converting enzyme inhibitors (ACE inhibitors) and angiotensin receptor blockers (ARBs) are widely prescribed medications. A 2014 recent Cochrane review of randomized controlled trials found no evidence of a difference in total mortality or cardiovascular outcomes (e.g. cardiovascular events) for ARBs as compared with ACE inhibitors for patients prescribed these medications for primary hypertension. However, whether this translates to non-study populations is not well understood. Is there a difference in all-cause mortality and cardiovascular outcomes for patients with diabetes and those who do not have diabetes using these medications irrespective of indication?
- Comparison for patients without diabetes
- All-cause 5-year mortality and (separately) cardiovascular events for patients without diabetes on ACE inhibitors.
- Numerator: Of those individuals in the denominator, number of individuals without diabetes on ACE inhibitors that experience death or a cardiovascular event within a 5-year period.
- Denominator: Number of individuals without diabetes on ACE inhibitors at a given point(s) in time.
- ii. All-cause 5-year mortality and (separately) cardiovascular events for patients without diabetes on ARBs.
- Numerator:Of those individuals in the denominator, number of individuals without diabetes on ARBs that experience a cardiovascular event or death within a 5-year period.
- Denominator: Number of individuals without diabetes on ARBs at a given point(s) in time.
- b) Comparison for patients with diabetes
- Total mortality and cardiovascular events for patients with diabetes on ACE inhibitors.
- Numerator: Of those individuals in the denominator, number of individuals with diabetes on ACE inhibitors that experience a cardiovascular event or death within a 5-year period.
- Denominator: Number of individuals with diabetes on ACE inhibitors at a given point(s) in time.
- Total mortality and cardiovascular outcomes for patients with diabetes on ARBs.
- Numerator: Of those individuals in the denominator, number of individuals with diabetes on ARBs that experience a cardiovascular event or death within 5-year period.
- Denominator: Number of individuals with diabetes on ARBs at a given point(s) in time.
Minimum sample size of 5,000 users of ACE inhibitors and 5,000 users of ARBs.
Data should follow users for at least five years.