Transcript www.hsri.org

STEP-BY-STEP ILLUSTRATION
FOR PREPARING A MORTALITY
REPORT
The Connecticut report is very “visual,” relying on a lot of graphs to
illustrate patterns and trends. This approach was used – as opposed
to a narrative dense format – to make it easier for general readers to
understand the data. Such an approach does, however, take a bit
more time to prepare, and requires setting up complimentary or
linked excel charts and graphs.
The report is published in powerpoint (only because I find that easier
to work in than word). It can be prepared with almost any publishing
software, as long as you can easily import tables and graphs.
The following pages present a step-by-step illustration of how to set
up and prepare the report. Obviously, available data and areas of
interest will determine the exact format and information that is
included in your report.
Health
and
ANNUAL
REPORT
Issued
NOVEMBER 2002
Mortality
This is the first of a series of semiannual reports on
trends and related information pertaining to the health
and quality of care received by individuals with mental
retardation served by the Connecticut State
Department of Mental Retardation. Future reports are
scheduled for March and September of each year.
September reports will focus on an analysis of annual
data, with a special emphasis on mortality trends.
March reports will focus on any significant or special
trends, new initiatives and important news and
information related to mortality and risk reduction.
For the Period July 1, 2001 to June 30, 2002
Overview of DMR
Mental retardation is a developmental disability that is present in about 1% of the Connecticut
population. In order for a person to be eligible for DMR services they must have significant deficits in
intellectual functioning and in adaptive behavior, both before the age of 18-yrs. DMR is also the lead
agency for the Birth to Three System in Connecticut. This system serves infants and toddlers with
developmental delays. Altogether, DMR assists almost 19,500 individuals and their families,
providing a broad array of services and supports.
THE PEOPLE
SERVED BY DMR
Includes Birth to Three children.
DMR provides or funds
residential supports for
6,621 people.
7,186 individuals living
at home without formal
residential support
This is simple demographic data that provides
a general overview of the population served by
the agency. It can be configured in any way that
makes sense for your agency.
62% of the people
we serve
Less than half
(38%) of the
people we serve
12,034
7,394
live in their own
homes or with
family without
residential support
live in
residential settings
Residential services for an
additional 773 people are
funded by other sources.
4,848 children living at
home and receiving only Birth
to Three services
as of 6/30/02
This data simply reflects the no. and percent of deaths
by where people lived (categories). You can configure
any way you want (usually it is helpful to distinguish between
persons who receive direct residential services from those
who do not and from those for whom another agency may
have responsibility (e.g., LTC).
Health
No. Deaths
34
6
5
53
27
49
4
178
An important component of the risk management systems present within DMR involve the analysis and
review of deaths to identify important patterns and trends that may help increase knowledge about risk
factors and provide information to guide system enhancements. Consequently, DMR collects
information on the death of all individuals served by the department. The following section provides a
general description of the results of this analysis for Fiscal Year 2002 (July 2001 through June 2002).
Percent
19%
3%
3%
30%
15%
28%
2%
100%
Type of Residential
Support
At Time of Death
Mortality and
Residence
Other
2%
During the 12 month time period between July 1,
2001 and June 30, 2002 a total of 178 out of the
19,500 individuals served by DMR passed away. As
can be seen in the graph to the right
approximately half died while being served in a
residential setting operated, funded or licensed
by DMR (blue section). The other half were living at
home (family home or independently), in a long-term
care facility (e.g., nursing home), or other non-DMR
setting . This general pattern is consistent with that
observed last fiscal year, although there was a slight
reduction in the relative percentage of deaths that
occurred in CLAs, Supported Living and Long-Term
Care facilities.
The average Death Rate* is expressed as the no. of
deaths per 1000 people served. It compares the
number of deaths to the number of persons served in
each type of setting (no. deaths /population X1000),
and continues to show a predictable pattern: In
general, the higher need for specialized care, the
higher the average rate of death.
This data represents death RATE, with table set up so that
it moves from lowest to highest. Rate is a good indicator
since it allows general comparisons that compensate for
differences in the population size. (If you use similar
groupings you will have CT and MA data as a benchmark.)
Excel chart set up as follows:
Crude
Rate
0.0044
0.0044
0.0087
0.0152
0.0297
0.1040
0.0151
LTC
28%
Campus
15%
No. Deaths per 1000 People
FY 2002
120
104.0
No./1000
4.416
4.444
8.696
15.199
29.670
104.034
15.094
100
80
60
40
20
29.7
15.2
4.4
4.4
Home
SL
8.7
0
CTH
CLA
Campus
Residential Setting
Home
19%
SL
3%
CTH
3%
CLA
30%
This graph shows the number
Mortality Rate by Where People Live
No. Deaths per 1000
People
Death Rate (per 1000) by Where People Lived
Population
Res Category
No. Deaths
EOFY
Total Pop
Home
34
7666
7700
SL
6
1344
1350
CTH
5
570
575
CLA
53
3434
3487
Campus
27
883
910
LTC
49
422
471
Other
4
261
265
Mortality Review ANNUAL REPORT
Mortality Trends
Excel chart set up as follows:
Res Category
Home
SL
CTH
CLA
Campus
LTC
Other
Total
&
September 2002
LTC
of people who died
for every 1000 people
served in each type of
setting. The settings
to the left tend to provide
less comprehensive care and support
than the settings to the right. This often
reflects the level of disability and
specialized care needs of the people
who generally live in each type of
setting.
For example, persons living in LTC
(nursing homes) tend to be older
than other people served by DMR,
and, usually went to a nursing home
because they needed skilled nursing
care. Their death rate is much higher
than for other people served by DMR.
* In this report we use the term “average death rate” to reflect what is more commonly referred to as the “crude” death rate in mortality and
epidemiological research.
-2-
Home = people living with families or independently; SL = supported living; CTH= Community Training Home (foster care); CLA=
Community Living Arrangement (group home); Campus = regional centers and STS (institutions); LTC=Long Term Care (Nursing
Homes, SNFs, etc.); Other = Everything else.
This data simply compares the past two years in terms of the
total no. deaths and the death rates. As time goes on,
additional years (reflecting trends) will be included.
Health
&
Mortality Review ANNUAL REPORT
September 2002
Mortality Rate
No. Deaths
Comparison: FY01 and FY02
Excel chart set up as follows:
The two graphs to the right
Percentage of Deaths
FY01
FY02
17%
19%
5%
3%
3%
3%
31%
30%
15%
15%
26%
28%
2%
2%
100%
100%
No. per 1000
FY01
FY02
4.244
4.416
7.168
4.444
8.651
8.696
16.579
15.199
29.819
29.670
100.209
104.034
11.719
15.094
12.649
12.061
Rate by residential setting across the two years is shown
here. This is useful to identify any emerging trends.
FY01
100.2
Deaths by Gender
GENDER
Men
Women
Total
GENDER
Men
Women
Total
92
86
178
52%
48%
100%
Population
Total Population
EOFY
8167
8259
6413
6499
14580
14758
Rate
(No. Deaths
per 1000)
No. Deaths
Percent of
Deaths
92
52%
11.139
86
48%
13.233
178
100%
12.061
Crude
Rate
0.01114
0.01323
0.01206
104.0
100
Death
Rate/1000
11.139
13.233
12.061
FY01
FY02
death rate (the number
deaths per 1000 persons served)
for fiscal year 2002 with that for last fiscal
year (FY2001) by type of residential setting.
80
Small differences can be seen, with the rate
decreasing for persons living in CLAs (group
homes) and in Campus settings (STS and
regional centers). The most pronounced
decrease occurred for persons receiving
Supported Living services.
60
40
29.8 29.7
16.6 15.2
20
4.2
4.4
Uses same excel chart as above
Excel chart set up as follows:
12.061
This graph compares the
7.2
4.4
8.7
8.7
0
Comparison of death rate by gender is a common (and almost
expected) type of analysis. It is useful since there is a lot of
data of differential rates (e.g., CDC). Make sure you also look
at age – higher death rates for women may be reflective of the
fact that their average age is higher.
12.649
FY02
120
Home
Percent of
No. Deaths Deaths
178
Comparison of FY01 and FY02
Mortality Rate by Where People Live
No. Deaths per 1000 People
No. Deaths
FY01
FY02
31
34
10
6
5
5
57
53
28
27
48
49
3
4
182
178
182
compare the number of deaths
within the population served by
DMR and the average death rate for
fiscal years 2001 and 2002. As
can be seen, FY02 experienced a
slight decrease in both measures.
Comparison: FY01 and FY02
Res Category
Home
SL
CTH
CLA
Campus
LTC
Other
Total
No. Deaths per 1000
Comparison: FY01 and FY02
SL
CTH
CLA
Campus
FY01
FY02
LTC
Slight increases in the mortality
rate occurred for persons living in
Long-Term-Care facilities and
their family homes
Caution must be exercised in reviewing this data since the actual number of deaths in each of these settings
was relatively small. The differences across this time period are therefore most likely not statistically
significant.
Gender and Mortality
During Fiscal Year 2002 a little over half (52%) of the 178 individuals who passed away were men.
However, DMR serves more men than women. The No. Deaths per 1000 people served shows that women
tend to have a higher death rate. It is important to note, however, that the average age of women served by
DMR is greater than the age of men, with almost two times as many females over the age of 85-yrs than
males. Thus, a higher death rate for women would be expected since they are, as a group, older than the
men served by the department.
GENDER
GENDER
Men
Women
Total
No.
Deaths
Percent of
Deaths
No. Deaths per 1000
FY 2002
Rate
(No. Deaths
per 1000)
92
52%
11.139
86
48%
13.233
178
100%
12.061
13.233
11.139
Men
-3-
Women
FY02: MORTALITY RATE BY AGE RANGE
AGE RANGE
(in
No. Deaths
years)
0-9
10-19
20-29
30-39
40-49
50-59
60-69
70-79
80-89
90+
Total
AGE RANGE
years)
0-9
10-19
20-29
30-39
40-49
50-59
60-69
70-79
80-89
90+
Total
Percent of Population
Deaths
EOFY
9
4
7
20
28
35
33
21
16
5
178
5%
2%
4%
11%
16%
20%
19%
12%
9%
3%
100%
No. Deaths
Percent of
Deaths
9
4
7
20
28
35
33
21
16
5
178
5%
2%
4%
11%
16%
20%
19%
12%
9%
3%
100%
(in
1357
2417
2463
2598
2622
1729
796
421
157
20
14580
Death
Rate/1000
6.589
1.652
2.834
7.639
10.566
19.841
39.807
47.511
92.486
200.000
12.061
Total
Population
1366
2421
2470
2618
2650
1764
829
442
173
25
14758
Crude Rate
0.00659
0.00165
0.00283
0.00764
0.01057
0.01984
0.03981
0.04751
0.09249
0.20000
0.01206
Death
Rate/1000
6.589
1.652
2.834
7.639
10.566
19.841
39.807
47.511
92.486
200.000
12.061
This table is a summary
of above data used for
direct import into the
report.
This analysis may not be necessary. It was included due to
erroneous reporting by the Hartford Courant and to illustrate
that there is NOT any adverse risk for people living in CLAs.
If you decide to include something like this, it will be
necessary to compute the average age of persons in each
type of residential setting and the average age of death for
people in those settings.
September 2002
Mortality Rate by Age Range
No. Deaths per 1000 People Served
FY 2002
Age and Mortality
The relationship between
age and mortality shows
the expected trend, with the
mortality rate increasing as
people served by DMR get
250
No. Deaths per 1000
Excel chart set up as follows:
Health & Mortality Review ANNUAL
REPORT
older. As seen here,
at around 70 -yrs of age
there is a dramatic rise in
mortality,
again, in line with
expectations and the trends
observed in
the general population.
AGE RANGE
(in years)
0-9
10-19
20-29
30-39
40-49
50-59
60-69
70-79
80-89
90+
Total
No.
Deaths
9
4
7
20
28
35
33
21
16
5
178
200.0
200
150
92.5
100
39.8
50
6.6
1.7
2.8
7.6
0-9
10-19
20-29
30-39
10.6
19.8
40-49
50-59
47.5
0
60-69
70-79
80-89
90+
AGE RANGE
Percent of
Deaths
5%
2%
4%
11%
16%
20%
19%
12%
9%
3%
100%
Individuals living at home
(especially those living with their
family) are younger than the other
persons served by DMR. They
also have a much younger average
age at death. The oldest group
served by DMR are living in LTC
facilities. They have the highest
average age at death.
Death
Rate/1000
6.589
1.652
2.834
7.639
10.566
19.841
39.807
47.511
92.486
200.000
12.061
As can be seen in this graph
there is a relatively strong relationship
between the average age of the
population living in each type of
residential setting and their average
age at death. The largest difference
between the two variables exists in
CLAs, where the average age of
death is 13-yrs higher than the
average age of the population living in
this type of setting.
Comparison of Average Age & Average
Age at Death by Residential Setting
FY 2002
80
68
70
64
58
60
Age (in Years)
Age is probably the most common type of analysis for mortality since
it is the strongest predictor of death and there are numerous
benchmarks you can look at (including data specific to a given
state). Data reflects a basic frequency distribution, converted to
rates (no. divided by total no. served in the age range). You can
expand or compress the age ranges to best reflect the population
you serve or specific program/service categories.
50
45
51
48
42
40
53
45
39
32
30
24
20
10
0
Home
SL
CTH
CLA Campus
Residential Setting
Ave Age
-4-
Ave Age at Death
LTC
Excel chart set up as follows:
FY 2002 - MR LEVEL and MORTALITY
LEVEL OF MR
No MR or ND
Mild
Moderate
Severe
Profound
Total
No. Deaths
18
42
36
40
42
178
Percent of
Deaths
10%
24%
20%
22%
24%
100%
Population
EOFY
1951
4781
4230
2005
1613
14580
Total
Population
1969
4823
4266
2045
1655
14758
Crude
Rate
0.009
0.009
0.008
0.020
0.025
0.012
Death
Rate/1000
9.142
8.708
8.439
19.560
25.378
12.061
The following sections of the report require additional data that may
not be readily available without a specific database associated with
the process of mortality review. You do have data on whether the
death was “natural” or “accidental” as well as a few other very
general categories. (Recommend you decide what type of
information you want to report on and then incorporate into your
reporting format and process.)
This data reflects deaths associated with hospice (and
therefore “expected.”
Health
&
Mortality Review ANNUAL REPORT
September 2002
Level of Disability and Mortality
In addition to age and gender, the
level of mental retardation is another
factor that affects a person’s life
expectancy. Persons with more
severe levels of disability typically
have many co-morbid conditions (other
medical diagnoses such as epilepsy,
cerebral palsy, etc.), including mobility
and eating impairments. These
disabilities have a significant effect on
morbidity (illness) and mortality.
As can be seen in this graph,
the relationship between level of
mental retardation and mortality shows
the expected trend. Persons with the
most significant levels of mental
retardation (severe and profound)
have a much higher rate of mortality.
Mortality Rate by Level of Disability
for Persons Served by DMR
FY 2002
30
26.04
No. Deaths per 1000
This data simply reflects the relationship between MR level
and mortality rate. You can do something very similar with
any data re: client characteristics as long as you can group
your population according to the variable (e.g., ICAP scores
on selected functional or diagnostic categories/levels).
The average rate is illustrated by the dashed line (to allow
easy identification of levels above and below the average).
25
19.95
20
15
Overall average rate = 12.209
9.23
10
8.78
8.51
Mild
Moderate
5
0
No MR or
ND
Severe
Profound
Level of Mental Retardation
No MR or ND category Includes children receiving DMR services through
the Birth-to-Three system who are too young to test for mental retardation
and adults for whom the DMR has limited responsibility under the Federal
Nursing Home Reform Act (OBRA 87) who do not have mental retardation.
It may include some DMR clients who were DMR clients prior to
Connecticut’s current statutory definition of mental retardation.
During FY 2002 (July 1, 2001 to June 30, 2002) 123 cases were formally reviewed by DMR Mortality Review Committees.
The information presented in the next section summarizes ONLY those deaths that were reviewed and will therefore be
different from the numbers discussed in the preceding section. Information regarding FY02 reviews will be presented for
ALL CASES REVIEWED and for only those deaths that OCCURRED DURING FY02.
DMR Mortality Review
DMR policy establishes formal mechanisms for the careful review of consumer deaths by local regional
Mortality Review Committees and a central Medical Quality Assurance Board. This latter entity, modified by
the Governor’s Executive Order No. 25, is now called the Independent Mortality Review Board (IMRB) and
includes representation from a number of outside agencies as well as a consultant physician. During FY02
a total of 123 cases were reviewed by these local committees and the central IMRB. Of these, 58 cases
were referred by local committees to the IMRB, and an additional 14 (11%) cases of the 65 closed at the
local level were reviewed centrally as a quality assurance audit.
Of the 123 cases that were reviewed, 41 represented deaths that occurred during FY02. Information
regarding these deaths is summarized separately below.
IMPORTANT FINDINGS From Mortality Reviews:
Community Hospice Support is routinely provided for persons served by DMR in all
Excel chart set up as follows:
types of residential settings, including regional centers and STS, CLAs, CTHs, and for individuals
receiving supported living services when death is anticipated, usually due to a terminal illness.
ALL CASES REVIEWED: Hospice support was provided in 24 of the 123 cases reviewed (20%)
FY02 DEATHS ONLY: Hospice support was provided for 14 of the 41 individuals who died (34%)
HOSPICE
All Reviews
FY02 Deaths Only
No.
Percent
Reviewed
24
20%
123
14
34%
41
-5-
All of this data had to be reported in two different ways since the
mortality review process is always a bit behind the actual fiscal
year, and, given the heightened attention to death that year, it
was important to present info on both the total no. of reviews as
well as only those deaths that occurred in that year. Most of the
data is self-explanatory. The categories chosen for inclusion
tended to be “hot button” issues where real objective information
can “calm the storm.”
Health
&
Mortality Review ANNUAL REPORT
September 2002
Autopsies are performed by the Office of the Chief Medical Examiner for those cases in which the
OCME accepts jurisdiction or by private hospitals when DMR requests and the family consents to the
autopsy.
ALL CASES REVIEWED: Of the 123 individuals reviewed, autopsies had been requested for 48
(or 39% of the sample), and consent was obtained and autopsies performed for 26 (21% of the
sample). The OCME accepted jurisdiction and performed autopsies for 15 of these cases, and
private autopsies were conducted for 11.
FY02 DEATHS ONLY: Of the 41 deaths that occurred during FY02, autopsies were requested for
22 (54%). A total of 8 autopsies were performed (20%), 5 of which were conducted by the OCME.
Excel charts set up as follows:
AUTOPSIES
All Reviews
FY02 Deaths Only
PREDICABILITY
All Reviews
FY02 Deaths Only
DNR
All Reviews
FY02 Deaths Only
No. Req % Req
48
39%
23
56%
No. Antic &
Rel to Dx
79
23
No.
71
15
No.
OCME
15
5
Unantic &
Rel to Dx
%
64%
29
56%
13
%
58%
37%
No.
Private % Cmpltd Reviewed
11
21%
123
3
20%
41
Unantic &
Not Rel to
Dx
%
24%
15
32%
5
%
Reviewed
12%
123
12%
41
No.
Meeting
Criteria Reviewed
67
123
15
41
Special Note: A recent report by the Columbus Organization found that the average rate of autopsy for
persons served by those state MR/DD agencies they surveyed was 11.7%. This compares to the 20-21%
rate noted above for cases reviewed by mortality review committees in Connecticut during FY02.
Predictability.
ALL CASES REVIEWED: In 64% of the cases reviewed (n=79), the death was anticipated and
related to the diagnosis. In another 24% of the cases (n=29) the death was not anticipated, but
was directly related to the existing diagnosis. In 12 % (n=15) the death was not anticipated and
not related to the diagnosis, as follows:
1 – heart anomaly
2 – asphyxia (drowning)
3 – cardiovascular disease
1 – subdural hematoma
1 – adverse drug reaction
1 – stroke
2 – pulmonary embolism (1 following surgery)
1 – pneumonia
2 – inhalation of food
1 – cause undetermined
by OCME
FY02 DEATHS ONLY: Of the 41 deaths reviewed that occurred in FY02, 56% (n=23) were
anticipated and related to the known diagnosis, 32% (n=13) were not anticipated but were related to
the existing diagnosis, and 12% (n=5) were not anticipated and not related to the diagnosis, as follows
(also
included above):
1 – cardiovascular disease
1 – stroke
1 – adverse drug reaction
1 – pulmonary embolism following orthopedic surgery
1 – cause undetermined by OCME
DNR.
Risk
All Reviews
FY02 Deaths Only
No. Mob
Imp
54
18
%
44%
44%
No. Asst
Eating
62
10
Total with
Risk
%
Factor
50%
116
24%
28
Do Not Resuscitate (DNR) orders are sometimes utilized when individuals reach the terminal
phase of an illness. DMR has an established policy that includes specific criteria that must be met along
with a review process for all DNR orders issued for persons served by the department.
%
Reviewed
94%
123
68%
41
ALL CASES REVIEWED: Of the 123 cases reviewed, 71 people (or 58%) had DNR orders, indicating
that their condition was terminal. Of these, 67 were formally reviewed by DMR. For the remaining four
individuals, DMR was not notified as required by policy, but in all cases the DNR was appropriate and
would have met established criteria. Of these four, two occurred at a LTC facility, one at an acute care
hospital , and the fourth at a Hospice facility. All facilities received additional training regarding
required notification to DMR.
FY02 DEATHS ONLY: Of the 41 deaths that occurred in FY02, 15 had DNR orders (37%). All met
DMR policy requirements (met criteria, and both notification and review took place as required).
Risk.
Mobility impairments and need for special assistance eating are two factors that place individuals
at significantly higher risk of death. The mortality review process therefore looks carefully at the presence
of these two personal characteristics.
Note: your mortality review process will need to be structured
to generate specific kinds of data and conclusions in order to
provide this type of data for a report.
ALL CASES REVIEWED: Of the 123 individuals reviewed, 54 – or 44% were non-ambulatory. 62,
or 50%, were not able to eat independently.
FY02 DEATHS ONLY: Of the 41 FY02 deaths reviewed, 18 (44%) had mobility impairments (nonambulatory) and 10 (24%) were not able to eat independently.
-6-
Context is similar to the data you collect. (Recommend you
carefully review the reliability of the reporting, especially to see if
later review causes changes to be made.)
Health
&
Mortality Review ANNUAL REPORT
September 2002
SUMMARY
Context.
Excel chart set up as follows:
Context
All Reviews
FY02 Deaths Only
Nat Cause
115
39
%
Accident
93%
6
95%
1
%
ND
5%
2%
%
1
1
Hom
1%
2%
%
1
0
1%
Reviewed
123
41
ALL DEATHS REVIEWED: The vast majority – over 90% of all deaths reviewed were classified as due to Natural
Causes. Six (6)deaths were associated with an Accident.
Of these, 2 were related to choking, 2 were related to
drowning, and 2 appear to be related to a fall. One case
was a Homicide and in one case the context was not able to
be determined by the OCME.
FY02 DEATHS ONLY: 39 deaths – or 95% - of the 41
reviewed were related to natural causes. 1 death was
accidental and 1 was not able to be determined by the
OCME. The accidental death was related to a fall.
Deaths that Occurred
and Were Reviewed
between 7/1/01 &
6/30/02
•
•
•
Neglect.
Neglect data reflects whether or not there was an allegation of
abuse or neglect within 6 mo. of the death. Therefore, it was
important to also note whether or not the neglect was related to
the death (which only an investigation or mortality review
process can determine)
Excel chart set up as follows:
Neglect
Name
Doe
Smith
Jones
Apple
Orange
Grape, Gilbert
Pear
Peach
Banana
Cherry
Neglect Sub
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No.
No.
Neg Cause of Death
No
Yes
Yes
No
No
Yes (OPA), No (DMR)
No
Yes
No
No
Action
ALL DEATHS REVIEWED: There were a total of 18
allegations of abuse or neglect that occurred within 6
months of death for the cases reviewed. Of these, 2 were
not substantiated, 8 are still under investigation, and 8 were
substantiated. In 4 of these latter cases, the neglect
appeared to be related to the cause of death, as follows:
2 - asphyxia resulting from drowning (private CLAs)
1- anoxia, associated with nursing failure to properly
assess (LTC)
1 - anoxia resulting from choking on food (private day
program)
Enforcement action was taken in 3 of the 4 cases and
included: 2 dismissals from service by the provider with
arrest by police and 1 citation with monetary fine by DPH
(1). In the fourth case there were inconsistent
findings regarding the culpability of the involved staff
member. In all four instances family members were
notified of findings.
FY02 DEATHS ONLY: Of the 41 deaths that occurred in
FY02 there were a total of 8 that included an allegation of
abuse or neglect within 6-months of death. Of these, 1 was
not substantiated, 5 are still under investigation, and in two
cases the neglect was substantiated. In both of these latter
two cases it was not possible to determine if the neglect was
the direct cause of the deaths. Both cases involved nursing
personnel where enforcement action included appropriate
reporting to the Department of Public Health and Nursing
Board.
Dismissed, Arrested
Dismissed, Arrested
Rep to Nurs Brd, Dismissed, Reinstated
Rep to DPH for invest
No discp action
Rep to Nurs Brd, Dismissed
Rep to DPH, Substantiated, facility cited & $ fine
34% of the people had
Hospice support.
20% had an Autopsy.
56% of the deaths were
Anticipated and
related to the existing
diagnosis. In 12% the
death was not anticipated
and not related to
the existing diagnosis.
• 37% had a DNR order. All
met DMR criteria.
•
44% of the people could
Not Walk (i.e., were
non-ambulatory).
•
24% could Not Eat
without assistance.
•
95% of all the deaths
reviewed were due to
Natural causes.
1 death was classified as
Accidental.
• 2 cases involved
Neglect that was
•
substantiated. In both cases it
was not possible to determine
if the neglect was related to
the cause of death.
-7-
Location at time of death may be important to look at,
particularly if questions get raised about the care being
provided within programs operated or funded by your
agency. It can be useful to show that a majority of
deaths take place within hospitals and LTC (as would
be expected). However, to do this, you will need to
capture that data either on the IR form or through
mortality review.
Health
LOCATION
Hospital
Hospice Facility
SNF
Conv Home
DMR Campus
STS HCU
CLA
CTH
Sup Lvng
Family Home
Other Community
ND
Total
No.
Percent
As can be seen in this graph
over 60% of the individuals reviewed by the mortality review
committee in FY02 passed away outside of a DMR - operated or
funded residential setting. Most died in the hospital or long term
care facility. The table below shows both the number of individuals
who died by location as well as the relative percentage by location.
51
2
15
1
8
5
21
1
5
1
1
12
123
41%
2%
12%
1%
7%
4%
17%
1%
4%
1%
1%
10%
100%
FY02:
LOCATION
DMR Setting
Other
Total
No.
40
71
111
Mortality Review ANNUAL REPORT
Location at Time of Death
Excel chart set up as follows:
Location at Time of Death
&
September 2002
Percent
36%
64%
100%
Leading causes of death is VERY important. IT is also a
variable that has ample benchmarks. You can use SD
data as well as national data (for both the general
population and, given published mortality reports, for the
MR/DD population as well – just make sure there is an
ability to directly compare DD populations since some
states only serve MR, some report only on adults, etc.).
LOCATION
No.
Percent
Hospital
Hospice Facility
SNF
Conv Home
DMR Campus
STS HCU
CLA
CTH
Sup Lvng
Family Home
Other Community
Hosp ER
Total
51
2
15
1
8
5
21
1
5
1
1
12
123
41%
2%
12%
1%
7%
4%
17%
1%
4%
1%
1%
10%
100%
DMR
Setting
36%
Other
64%
Where People Died
FY 2002 Mortality Reviews
LEADING CAUSES
OF DEATH
A review of data from Connecticut and two other New
England states suggests that the leading causes of death
for people with mental retardation are somewhat different than for the general population. Heart disease is
the no. 1 cause of death – for all groups. However, unlike the general population, deaths due to
respiratory conditions are the second leading cause of death for individuals served by DMR. This
is expected due to the high percentage of deaths for persons with severe and profound mental retardation
and the high incidence of co-morbid conditions in that group, including conditions such as cerebral palsy,
dysphagia, gastro-esophageal disorders, all of which carry a heightened risk of aspiration pneumonia. It
should be noted that increasing age is an important factor that increases risk for aspiration pneumonia as
documented in the National Vital Statistics Report published by the CDC.1 This report states that a major
cause of death “concentrated among the elderly, is a pneumonia resulting from aspirating materials into the
lungs.”
Diseases of the nervous system are the third leading cause of death for DMR consumers.
These include Alzheimer’s Disease – which has a very high incidence in people with Down Syndrome and Seizure Disorders, again a condition that has a much higher incidence in people with mental
retardation. Interestingly, deaths due to accidents are much lower for people with mental
retardation than for the general U.S. or Connecticut population. Deaths due to injuries or accidents are the
5th leading cause of death in the general population , but are only the 8th highest cause of death for people
reviewed by DMR’s mortality review committees.
Leading Causes of Death
RANK
Connecticut DMR
Connecticut
U.S.
1
Heart Disease
Heart Disease
Heart Disease
2
Respiratory Disorders
Cancer
Cancer
Nervous System
Disorders
Respiratory Disorders
Cerebrovascular Diseases
Cancer
Accidents
3
4
(incl Stroke)
Chronic Respiratory
Diseases
Health
Benchmarking is becoming more and more critical since it
provides a means to assess whether or not your state is
“typical” or an outlier. It will be important to scan the web and
stay in touch with other states to get copies of mortality data
and reports as they become available. Be very CAREFUL,
however – not all data is comparable. CT and MA have similar
systems – communicate a lot – and therefore can usually use
one another as a benchmark. (Check with Wanda to see if she
was able to get the report from Gerry Morrisey.)
&
Mortality Review ANNUAL REPORT
September 2002
BENCHMARKS
While there is a dearth of objective information regarding mortality in persons with mental
retardation being served by state agencies from across the country, this section will provide
comparative analysis when appropriate benchmarks do become available.
Massachusetts DMR
The Massachusetts Department of Mental Retardation has recently enhanced and expanded its mortality
reporting requirements and has issued an annual report. This 2000 Mortality Report was prepared by the
University of Massachusetts Medical School/Shriver, Center for Developmental Disabilities Evaluation and
Research2. The report covers the calendar year January 1 through December 31, 2000. Mortality statistics
pertaining to persons 18-years and older served by DMR were analyzed according to a number of variables not
dissimilar from many of those contained in the first part of this report. Consequently, it is possible to use some
of the Massachusetts data for comparative purposes. It should be noted that the Massachusetts DMR system,
although larger, is very similar to Connecticut’s (e.g., population served, type of services and supports,
organization). However, there are differences in reporting requirements, age limits, and and categorization of
service types. It is therefore important that readers exercise caution when reviewing comparative information.
Overall Death Rate
Massachusetts DMR - 18+
Mortality and AGE
AGE RANGE
18-19
20-29
30-39
40-49
50-59
60-69
70-79
80-89
90+
Total
RESIDENCE TYPE
Comm Resid
DMR Campus
LTC
Home
Other
Total
(in years)
No. Deaths
0
7
20
28
35
33
21
16
5
165
0%
4%
12%
17%
21%
20%
13%
10%
3%
100%
MA Total MA Percent of
Population
Population
10126
42%
1260
5%
1199
5%
10693
45%
643
3%
23921
100%
Population
EOFY
580
2463
2598
2622
1729
796
421
157
20
11386
Total
Population
MA No.
Deaths
110
47
112
48
5
322
580
2470
2618
2650
1764
829
442
173
25
11551
Crude Rate
0.00000
0.00283
0.00764
0.01057
0.01984
0.03981
0.04751
0.09249
0.20000
0.01428
12.1
8.7
Crude Death Rate = No. Deaths X 1000/Population + No. Deaths
Residential Analysis
13.5
COMPARISON OF AVERAGE DEATH RATES
Connecticut DMR v Massachusetts DMR
By Where People Live
Percent of
Deaths
13.5
Massachusetts
DMR - 18+
Crude Rate
8.7
12.1
14.3
14.3
Connecticut
DMR - 18+
Source
U.S. Population - all ages
Connecticut DMR - all ages
Connecticut DMR - 18+
General U.S. Population v People Served
by the MA & CT Mental Retardation Agencies
Connecticut
DMR - all
ages
Excel charts set up as follows:
COMPARISON OF AVERAGE DEATH RATES
U.S.
Population all ages
A comparison of the overall death rate for persons served
by the Connecticut DMR with similar rates for the general
population in the U.S. and the DMR population in
Massachusetts are presented in this graph.
The overall Connecticut DMR death rate of 12.1 deaths per
thousand people is higher than the rate of 8.7 deaths per
thousand people in the general population, as would be
expected due to the many health and functional
complications associated with disability and mental
retardation. A comparison of the Connecticut DMR with
Massachusetts DMR shows a slightly higher death rate in
Connecticut for the adult population (people older than 18yrs of age.) of 0.8 deaths per thousand people served.
This difference does not appear to be significant and may
be a reflection of the aforementioned differences in the
populations being served.
Death
Rate/1000
0.000
2.834
7.639
10.566
19.841
39.807
47.511
92.486
200.000
14.284
CT Total
MA CDR
Population
10.9
5310
37.3
910
93.4
471
4.5
7700
7.8
367
13.5
14758
CT Percent
of
Population
36%
6%
3%
52%
2%
100%
Comm
Resid
DMR
Campus
LTC
Home
Other
Total
MA CDR
10.9
37.3
93.4
4.5
7.8
13.5
CT CRD
11.7
29.7
104.0
4.4
16.3
12.1
CT No.
Deaths
62
27
49
34
6
178
-9CT CRD
11.7
29.7
104.0
4.4
16.3
12.1
A comparison of average death rates
by where people live is presented
here.
The general pattern for
rates by type of setting is quite similar
across the two states, with the
exception of the “Other” category.
This is most likely a reflection of
differences in the populations
included in this cluster.
Death rates in DMR would
therefore appear to be very
consistent with an available
benchmark as reported in
Massachusetts.
This page is purely optional. Including it shows that the
department is paying attention to what is happening nationally,
and in a way “forces” staff to take time to scan national
literature. If you conduct any statistical studies (e.g.,
relationship between ICAP and mortality, or other types of
incidents and mortality) it would be most beneficial to
summarize findings in this section of the report.
Health
&
Mortality Review ANNUAL REPORT
September 2002
RESEARCH & REPORTS OF
INTEREST
This section will report on selected research, reviews, and other information from Connecticut and around
the country that is related to mortality and health care in mental retardation and developmental disabilities
systems.
Connecticut DMR Independent Study on Mortality
The Connecticut DMR retained the services of two outside consultants to conduct a comprehensive analysis of
mortality and basic demographic trends from 1997 to 2002 within the population of individuals served by DMR.
The study was designed to provide:
•
Descriptive Overview of People Served by DMR
•
Predictive Mortality Analysis
•
Cross-sectional Analysis of People Served
•
Longitudinal Analysis (Changes over Time)
Using sophisticated statistical procedures the study authors found that:
1.
Changes in mortality rates over time are not significant
2.
As expected, mortality is highly related to client age
3.
Women served by DMR are older than men, and hence have a higher mortality rate
4.
Increased levels of disability are inter-related and correlated with higher risk of mortality
5.
The strongest predictors of mortality are age, mobility status, and amount of supervision
provided
6.
The “aging in place” phenomenon is leading to increased risk of mortality since
individuals served by DMR are becoming older and more disabled over time.
Copies of the report3 and a graphical summary can be obtained by contacting:
DMR Strategic Leadership Center
860-418-6163 or [email protected]
California Study of National Mortality Review Systems
The Columbus Organization conducted a survey of national mortality review practices in MR/DD systems for the
California DDS4. Survey findings indicate that:
•
The majority of states require reporting of deaths for persons served by state DD agencies
at both the local and statewide level.
•
In most instances the determination to perform an autopsy is based upon the unique
circumstances of each case, with an average of 11.7% of all cases having an autopsy.
•
About half of the states use a set of standardized criteria to review deaths.
•
The majority of states have established databases to track mortality information.
The Columbus report was published in May of 2002. Copies can be obtained by contacting Columbus at
800-229-5116.
References
1 Minino, M.P.H., Arialdi, M. and Smith, Ed., S.B., CDC National Vital Statistics Reports National Vital Statistics System,
Deaths: Preliminary Data for 2000, Volume 49, Number 12, October 9, 2001.
2 2000 Mortality Report: A Report on DMR Deaths January 1 – December 31, 2000. Prepared for the Massachusetts
Department of Mental Retardation by the Center for Developmental Disabilities Evaluation and Research at the
University of Massachusetts Medical School/Shriver. March 4, 2002.
3 Gruman, C. & Fenster, J. A Report to the Department of Mental Retardation: 1996 through 2002 Data Overview
Completed: April 2002.
4 The Columbus Organization. Mortality Review Survey: Survey of the States. Submitted to the California Department of
Developmental Services. May, 2002.
-10-
This section of the report is also optional. It does, however,
allow the agency to publicize new initiatives and efforts to
enhance services and systems that reduce risk (e.g.,
improving your IR and MR system).
Health
&
Mortality Review ANNUAL REPORT
September 2002
ENHANCEMENTS:
Executive Order No. 25
A number of important enhancements to the risk management and mortality review systems in DMR are being
implemented in response to Governor Rowland’s Executive Order No. 25. All of these changes are designed to
improve communication with families, assure that a rigorous and objective evaluation and review of circumstances
surrounding untimely deaths takes place,and to make sure that the review process is independent and free from
the potential for conflict of interest.
Some of these enhancements include:
Stronger Role for Investigations Unit
The Connecticut DMR has a unique relationship with the State Police that includes the assignment of a senior
Officer to oversee and manage the Investigations Unit. Two trained clinical nurse investigators have joined the
unit’s staff and are conducting preliminary screening on all deaths that occur in DMR operated or funded settings
to immediately assess the need for a complete A/N investigation. In addition, a Special Investigative Assistant has
been appointed to oversee and monitor investigations conducted within the private sector.
New Independent Mortality Review Board
The Medical Quality Assurance Board has been transformed into a new Independent Mortality Review Board that
increases outside representation. The Chairperson was appointed by the Commissioner of DMR, in consultation
with the Director of the Office of Protection and Advocacy (OPA). The independent medical professional
(physician)and an independent representative from a private sector agency were jointly appointed by the DMR
Commissioner and OPA Director. In addition, OPA now has two members. The new IMRB began meeting in
March, 2002.
Increased Communication with OPA
The department is notifying the Executive Director of the Office of Protection and Advocacy of all deaths that occur
for persons served by DMR. The Director may request an expedited review by the IMRB, or, may direct that an
abuse/neglect investigation be initiated for any case.
Consistent Notification of Families
New policies and procedures have been implemented to assure that families and guardians are consistently
notified of all deaths and the results of investigations and mortality reviews. Families are provided with an
opportunity to meet with DMR personnel to review all findings.
Posting of Licensing Inspection Reports
The department is now requiring visible notice to consumers, families and guardians that the results of DMR
licensing inspections are available for review. In addition, DMR is posting summary reports of inspections on the
DMR website in order to make access to the information much easier and more widely available to the public.
Results of licensing inspections can be viewed at www.dmr.state.ct.us/license.htm.
The Next Health and Mortality Review UPDATE
Will be issued in March of 2003.
For more information or to contact DMR please visit us at
www.dmr.state.ct.us
Prepared by:
Steven Staugaitis, Director, Strategic Leadership Center
Marcia Noll, Director, Health and Clinical Services
-11 -