Transcript Evaluation of the Telephone Nurse Monitoring Program

Increased treatment completion for latent
TB infection with the Telephone Nurse
Monitoring Program (TNMP)
Michelle Macaraig, DrPH, MPH
Assistant Director for Strategic Planning and Program Evaluation
Bureau of Tuberculosis Control
New York City Department of Health and Mental Hygiene
Evaluation of contacts to TB cases in NYC
• Contacts to TB patients are
screened and evaluated at
–
–
–
–
Field (TB test)
TB chest centers
Hospitals
Other providers
• DOHMH case managers ensure
proper evaluation and follow-up
of all contacts
• Evaluate approximately 4,000 contacts annually
– Over 500 start treatment for latent TB infection of which 50%
are treated at a TB chest center
Treatment of contacts with latent TB
infection in TB chest centers
• Treatment for LTBI is nine months on self-administered
isoniazid (INH)
• Treatment completion among contacts with LTBI in NYC is
consistently below 70%
• Limited success to increasing treatment completion with
– Directly observed therapy for LTBI
– Nurse home visits
– Incentives
• Barriers to completing treatment include
– Length of treatment (9 months)
– Required monthly clinic visits
– Lengthy waiting times at TB chest centers
LTBI treatment initiation and percent completion,
NYC 2004-2007
Target: 79%
700
68%
Number started treatment
65%
600
56%
59%
70
60
500
50
400
40
300
30
200
20
100
10
0
0
2005
2006
2007
2008
Years
Number of contacts started treatment
Percent of contacts completed treatment
Percent completed treatment
80
Why Telephone Nurse Monitoring
Program (TNMP)?
• Improve completion rate for treatment of latent
TB infection (LTBI)
• Leverage existing technology to facilitate
treatment adherence despite decreases in
resources
• Address barriers to treatment completion
• In 2006, piloted TNMP in one chest center and
found that treatment completion increased to
77%
What is TNMP?
• Program to engage and monitor eligible patients while
they are on treatment for LTBI
• Treatment monitoring
– First three months monitored by doctor and nurse, then
nurse at subsequent months
– Follow-up monitoring by telephone call interspersed with
in-person clinic visits
– Total of five clinic visits and four TNMP calls
• Medications are mailed to patient’s home one month at
a time after each successful TNMP call
Monitoring with TNMP
Initial
visit
Jan
1st
2nd
3rd
follow-up follow-up follow-up
visit
visit
visit
Feb
Mar
Apr
4th
follow-up
visit
May
1st
TNMP
call
2nd
TNMP
call
15 days 30 days
Jun
Jul
5th
follow-up
visit
Aug
4th
3rd
TNMP TNMP
call
call
15 days 30 days
Sep
Oct
Nov
Eligibility Criteria for TNMP
• Low risk for hepatic complications
– Baseline for liver function test
• Completed the three months of treatment
• Greater than or equal to 18 years old
• Able to communicate with nurse directly or with
translation through Language Line
• Read instructions on medication label
• Verified stable address
– Not homeless at the time of diagnosis
• Verified phone number
Preparing patients for TNMP call
• Schedule with the patient the dates and times of the call
following clinic visits
– Enter scheduled calls in the Electronic Medical Record
– Document calls in patient’s treatment card
• Discuss the process with the patients
– Expect calls within 15 minutes of agreed time
– Two call attempts will be made
– Establish security question or code to verify the nurse
reached the patient
• Educate on what to do in case of adverse reaction
Mail order medications
TNMP Historical Dates
Evaluation of TNMP
Study design
• Study population: eligible contacts who
started treatment for LTBI in 2008 in one
of the NYC DOHMH TB chest centers
• Excluded
– Died during treatment
– Developed active TB
– Treatment for LTBI was other than INH alone
Analysis
• Examined demographic and clinical characteristics of
contacts and their associated index case
• Compared the proportion of contacts enrolled in TNMP
versus contacts not enrolled in TNMP who completed
treatment
• Examined the effect of being enrolled in TNMP on
treatment completion while adjusting for other variables
• Pearson’s chi-square was used to compare proportions
• Poisson regression with robust variance estimator was
used for multivariate analysis
Results
Flow diagram of study population
Contacts started
treatment for LTBI
in 2008, n=912
Eligible contacts
n= 403 (44%)
Excluded
n=509 (56%)
• Less than 1 month on treatment
• Aged <18 years or no age
• TB disease
• Homeless
• Died during treatment
• Treatment other than INH
Excluded
Managed by private provider
n=158 (39%)
Treated at
DOHMH chest center
n=245 (61%)
TNMP
n=59 (24%)
No TNMP
n=186 (76%)
Index case
Contacts
Characteristics of contacts and their index case enrolled and not enrolled in
TNMP, 2008
Age <35
Race
Non-Hispanic White
Asian
Non-Hispanic Black
Hispanic
Unknown
Male
US born
Yes
No
Unknown
HIV
Positive
Negative
Unknown
Initial TB test type
TST
QFT-G
Positive TB test result
Chest x-ray result of the index case
Cavitary
Non-cavitary
Unknown
Culture result of the index case
Positive
Respiratory smear result of the index case
Positive
HIV status of the index case
Positive
Negative
Unknown
Close relation to the index case
Total
n= 245 (%)
105
(43)
TNMP
n=59
(%)
28
(47)
Not in TNMP
n=186
(%)
77
(41)
p-value
0.41
12
76
52
101
4
157
(5)
(31)
(21)
(41)
(2)
(64)
3
19
13
21
3
35
(5)
(32)
(22)
(36)
(5)
(59)
9
57
39
80
1
122
(5)
(31)
(21)
(43)
(1)
(66)
0.18
25
214
6
(10)
(87)
(2)
6
53
0
(10)
(90)
(0)
19
161
6
(10)
(87)
(3)
0.38
0
54
191
(0)
(22)
(78)
0
14
45
(0)
(24)
(76)
0
40
146
(0)
(22)
(78)
0.71
213
32
223
(87)
(13)
(91)
50
9
54
(85)
(15)
(92)
163
23
169
(88)
(12)
(91)
0.56
65
171
9
(27)
(70)
(4)
16
42
1
(27)
(71)
(2)
49
129
8
(26)
(69)
(4)
0.99
229
(93)
56
(95)
173
(93)
0.61
178
(73)
43
(73)
135
(73)
0.96
11
196
38
184
(4)
(80)
(16)
(75)
1
40
18
40
(2)
(68)
(31)
(68)
10
156
20
144
(5)
(84)
(11)
(77)
<0.001
0.38
0.88
0.14
Contacts who started treatment for LTBI enrolled and not
enrolled in TNMP by chest center, N=245
TNMP
No TNMP
Chest centers
N=245
N=59
%
N=186
%
Chest center 1
4
0
(0)
4
(100)
Chest center 2
4
0
(0)
4
(100)
Chest center 3
7
2
(29)
5
(71)
Chest center 4
21
5
(24)
16
(76)
Chest center 5
28
2
(7)
26
(93)
Chest center 6
34
12
(35)
22
(65)
Chest center 7
38
11
(29)
27
(71)
Chest center 8
49
12
(24)
37
(76)
Chest center 9
60
15
(25)
45
(75)
Number and percent of contacts enrolled and not
enrolled in TNMP by treatment outcome, N=245
TNMP
Not in TNMP
N=59
%
N=186
%
P-value
Completed
48
(81)
124
(67)
0.05
Not completed
11
(19)
58
(31)
Referent
Stop treatment
0
(0)
4
(2)
Not applicable
Effect of TNMP on treatment completion for LTBI, N=241
Completed treatment
N=172
(%)
Did not complete treatment
N=69
Enrolled in TNMP
Yes
48
(28)
11
No
124
(72)
58
Age
<35
70
(41)
35
>35
102
(59)
34
US born
Yes
21
(12)
4
No
148
(86)
64
Unknown
3
(2)
1
Race
Non-Hispanic White
8
(5)
4
Asian
57
(33)
19
Non-Hispanic Black
44
(26)
8
Hispanic
59
(34)
38
Unknown
4
(2)
0
Sex
Male
106
(62)
48
Female
66
(38)
21
Respiratory smear result
of the index case
Positive
124
(72)
51
Negative
48
(28)
18
HIV status of the index
case
Positive
4
(2)
7
Negative
142
(83)
51
Unknown
26
(15)
11
Relation to the index
case
Close
131
(76)
50
Casual
41
(24)
19
*Contacts who stopped treatment due to adverse reactions were excluded
Crude relative risk Adjusted relative risk
(%)
95% CI
95% CI
(16)
(84)
1.19 (1.02, 1.40)
ref
1.22 (1.04, 1.43)
ref
(51)
(49)
ref
1.12 (0.95, 1.32)
ref
1.07 (0.90, 1.26)
(6)
(93)
(1)
0.89 (0.49, 1.61)
ref
1.07 (0.61, 1.90)
(6)
(28)
(12)
(55)
(0)
ref
1.64 (1.40, 1.93)
1.18 (1.05, 1.32)
1.33 (1.17, 1.52)
1.50 (1.01, 2.24)
(70)
(30)
0.91 (0.78, 1.06)
ref
(74)
(26)
0.97 (0.82, 1.15)
ref
(10)
(74)
(16)
1.93 (0.86, 4.34)
ref
0.96 (0.76, 1.20)
(72)
(28)
1.06 (0.87, 1.28)
ref
ref
1.51 (1.28, 1.89)
1.09 (0.93, 1.28)
1.23 (1.04, 1.44)
1.40 (0.95, 2.08)
0.99 (0.96, 1.01)
ref
0.99 (0.96, 1.03)
Limitations and Strengths
• Limitations
– Contacts were not randomized to TNMP
– Although, characteristics of contacts enrolled and not enrolled in
the program were similar, there may be other factors not
examined that could have biased the results in either direction
– Thirty-nine percent of eligible contacts were excluded because
they were treated by an outside provider and could not be
offered TNMP
• Strengths
– Data on enrollment and follow-up of patients were available for
contacts in all chest centers
Conclusion
• Contacts enrolled in TNMP were more likely to complete
treatment compared to contacts not enrolled in TNMP
• Proportion of contacts enrolled in TNMP remained low
(less than 30%) despite efforts to expand to other chest
centers
• Increased enrollment in the program could improve
overall treatment completion among DOHMH chest
center patients
Challenge
• Patients change phone numbers or
address
• Language barriers
– Patients did not fully understand the process
when first accepted TNMP
– Calls took longer with interpreter
• More time for staff when multiple attempts
needed to reach patient
Benefits of TNMP
• Facilitates completion of treatment
• On average 45% fewer clinic visits = less
inconvenience for patient
• Can receive call at home, workplace or any
other place of patient choice
• Patient/nurse can initiate call on given
appointment date and time
• Can save provider time for higher priority
patients
Acknowledgments
•
•
•
•
•
•
Jennifer Pierre, DrPH
Shama Ahuja, PhD, MPH
Holly Anger, MPH
Errol Robinson
Cheryl Herbert
BTBC clinic staff
TNMP Evaluation of Chelsea
Patients
• Preference for Monthly follow-up
– 12 (86%) prefer the nurse to call
– 1 (7%) prefer to come to the clinic
– 1 (7%) says it depends on the situation
Overall LTBI Completions in the
Chest Centers
2004
2005
2006
2007
# Overall
LTBI starts
5,905
5,075
4,937
3,096
% Overall
LTBI
completions
47.1%
46.3%
42.0%
52.0%
LTBI Completion Rates:
Progress towards National Goals
80%
70%
60%
50%
40%
30%
20%
10%
0%
Overall
CDC Goals
2004
2005
2006
2007
75%
47%
46%
42%
52%