Epidemiology of Palliative Care/Stats

Question 1

Incidence

Answer 1

• Number of new events/cases of disease that develop in a population at risk during a specified time interval

= N new cases/N population at risk

Question 2

Prevalence

Answer 2

• Proportion of individuals in a population who have the disease at a specific moment in time and provides an estimate of the probability that an individual will be ill at a point in time

= N current cases/N population at a given point in time

Question 3

Sources of mortality estimates

Answer 3

1. Death registration systems (provide info, though not always complete), on COD for most high income countries
2. Sample death registration systems - register a sample of the pop and establish death rates that can then be extrapolated to the broader population. Useful in rural areas, countries like China and India (1/3 of world pop)
3. Epidemiological assessments - estimates of deaths for major diseases (cancer, HIV etc.) for countries in regions most affected. Case fatality rates in combination with incidence/prevalence used to estimate mortality rates
4. COD models - used to estimates deaths according to broad cause groups in regions with non-existent/incomplete mortality data

Question 4

Issues with mortality stats

Answer 4

• 1/3 of world pop lives in regions with complete civil registration systems
• Single cause of death may be used, when in reality a combination of comorbidities or health risks contributed to death (introduces biases)
• Coding and reporting systems not necessarily standardised (though movement towards using ICD)

Question 5

Life expectancy

Answer 5

• Varies considerably, from around 50 to 80 or above

- Social determinants and infant mortality rates are key

Question 6

Leading causes of death worldwide

Answer 6

1. IHD
2. Stroke
3. Lower resp infection
4. COPD
5. Diarrhoeal disease
6. HIV/AIDS
7. Lung CA
8. DM
9. Road injury
10. Prematurity

Question 7

Projections for causes of death

Answer 7

• Generally three kinds of estimates - baseline, pessimistic, and optimistic
• Projections largely dependent on access to preventative care, changes in disease mortality, social factors, etc.

Question 8

Location of death - Epi level

Answer 8

• Location of death generally only recorded in high-income nations
• Death in hospital is common in most high income countries (up to 50%), while death at home is more common in low income countries
• Most individuals indicate preference for care at home up to death

Research needed to match ‘preferred place of death’ to ‘actual location of death’

Question 9

Access to palliative care - epidemiological level

Answer 9

• Specialist palliative care is most available in high income countries
• In high income countries, more likely to have access when dying of a malignant process

In low income countries, the majority of those at EOL do not have access to specialist services, and where services exist, they reach only a very small number

Question 10

Factors influencing place of death

Answer 10

1. Illness factors
   - Non solid tumours (hospital)
   - Long length of disease (home)
   - Low functional status (home)
2. Individual factors
   - Ethnic minorities (hospital)
   - Good social conditions (home)
   - Patient preference (home)
3. Environmental factors
   - Use and intensity of home care (home)
   - Rural environment (home)
   - Available inpatient beds (hospital)
   - Greater hospital provision (hospital)
   - Prev hospital admission (hospital)
   - More social supports (living with someone, extended family support, marital status, caregiver’s pref) (home)
   - Historical trends (home)

Question 11

Methodological limitations of data related to symptoms

Answer 11

1. Symptom-related epidemiological data must be considered in the context of availability of management
   - 70% of patients being in pain not reflective of the fact that most patients have their pain adequately treated
2. Defined population
   - Care not to inappropriately extrapolate data (e.g. prognosis may be measured from time of dx, but not take into account disease stage)
3. Patient experience is personal and subjective
   - Every population has outliers and heterogeneity
4. Data accuracy is dependent upon accuracy of info communicated between subject and researcher
5. Symptom experience and burden of that symptom changes over time (point prevalence data may not always be applicable)
6. Symptom experience is multidimensional
7. Questions remain as to which symptoms/healthcare needs are most common and burdensome within particularly contexts

Question 12

Empidemiological data on symptoms over time

Answer 12

• Most patents with advanced, progressive, life-limiting disease often face a ‘core’ group of symptoms:

In the last 1-2 weeks of life with cancer, pooled prevalence by order of magnitude:

1. Fatigue
2. Pain
3. Lack of energy
4. Weakness
5. Loss of appetite

Question 13

Trajectories of functional decline toward end of life

Answer 13

1. Sudden death (high function, then sudden death with vertical decline in function)
2. Organ failure (stepwise decline in function over time, until death)
3. Terminal illness (high function, then steep decline in function toward death)
4. Frailty (Medium function, with gradual and slow decline to death)