Keywords
older adults - hypothyroidism - survey - thyroid hormone therapy - overtreatment -
age
Introduction
Hypothyroidism is the most common thyroid disorder in adults globally. Hashimoto's
thyroiditis has a female preponderance of 4:1; however, there are discrepancies in
the regions with different economic levels.[1] Population studies indicate that hypothyroidism prevalence increases with age, with
1.7% of adults over the age of 65 diagnosed with overt hypothyroidism.[2] The prevalence of subclinical hypothyroidism (defined as normal free T4 [FT4] and
elevated thyroid-stimulating hormone [TSH]) in adults over the age of 65 has been
quoted as high as 13.7% in North America when assuming 4.5 mIU/L as the upper limit
of normal for TSH.[1] There is compelling population data to support the concept that a “normal” TSH range,
which is the target for treatment, may be higher in older adults and associated with
longevity and better survival.[3]
[4]
[5] Lowering the TSH in these individuals may not be warranted and can pose a risk of
overtreatment, leading to detrimental consequences, especially in the older adults.[6]
[7] Risks include cardiac arrhythmias, increased cardiovascular mortality, osteoporosis,
and fractures.[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Despite the compelling evidence indicating an age-related increase in serum (TSH)
levels and the potential risks associated with overtreatment, current guidelines lack
specific recommendations for TSH goals tailored to different age groups among individuals
with hypothyroidism. Specific TSH ranges are currently only recommended for women
of childbearing age interested in conception and for pregnant women.[16] However, clarity is needed for the necessity for age-specific TSH targets in the
treatment of adults with hypothyroidism.[17]
[18]
[19]
[20] Furthermore, there is currently growing evidence against the routine treatment of
older adult patients with subclinical hypothyroidism.
What factors affect health care providers' decision-making when setting target TSH
treatment goals is still being determined. A study recently explored TSH goals in
hypothyroidism.[21] However, primary care providers are often the ones who prescribe thyroid hormone
preparations. Although another survey included primary care physicians, it was conducted
over two decades ago.[18]
Due to limited evidence about the benefits and risks of treatment on clinical outcomes,
it is still being determined whether treatment of subclinical hypothyroidism is necessary,
especially in older people.[22]
[23]
[24]
[25] Therefore, we conducted a multiregional physician survey to assess current clinical
practice patterns regarding TSH goals among a diverse group of clinicians and to elucidate
the factors influencing physicians' decision-making in managing adults with hypothyroidism.
Additionally, we aimed to investigate the variation in treatment strategies for older
patients with subclinical hypothyroidism, depending on select patient characteristics.
Materials and Methods
Target Population
The study population was defined by convenience sampling. In the absence of a single
regional endocrine society with a well-defined membership list that can represent
a study population, the target population was identified from a list of electronic
mail pooled from continuous professional development delegates, speakers, authors,
or members of various scientific groups or forums in different parts of the Middle
East and Africa (MEA) region. Consequently, several questions were added to the survey
to help define the demographic professional profiles of the respondents and their
practices similar to those of our previously published studies.[26]
[27]
[28]
[29] The study used a convenience sampling method, and therefore no sample size was calculated.
Also, the response rate could not be calculated as we have no information on the nonrespondents.
English was used being the main language used by the medical profession, although
some parts do admittedly use French. Due to the sample size restrictions and inhomogeneous
mismatch distribution between regions, we consciously refrained from making any subgroup
analysis.
Survey Management
A Web-based commercial survey management service (Survey Monkey, Palo Alto, California,
United States) was used. All participants received an initial email that explained
the rationale of the survey and what was required from the consented respondents,
followed by five subsequent reminder emails during the study period. Each message
included an explanation of the rationale and method of participation, full credentials
and contact details of the principal investigator, and a unique email-specific electronic
link to the questionnaire. The survey service automatically blocked repeated submissions
from the same IP (Internet Protocol) address. The survey Web site was open for the
12-month study period. Finally, survey responses were collected and stored electronically
for an anonymous analysis.
Survey Questionnaire
The questionnaire was adapted from two recently published studies with the same goals.[30]
[31] The questionnaire's contents were validated in the original studies and were used
in their original language. The descriptive part to characterize the responding population
has been established in many of our studies cited in the methodology section. However,
we have a small pilot test for the survey before we run out to the larger sample.
The text of the pooled survey instrument is provided in the online [Supplementary Material S1] (online only). Section I aimed to get insight into respondents' demographic and
professional profiles (8 questions). In addition, what is the number of people seen
in their practices? They were also asked how much time had elapsed since last diagnosing
a patient with subclinical hypothyroidism and how much time had elapsed since last
starting thyroxine treatment in a patient with subclinical hypothyroidism. Section
II included two survey items addressing the provider's decision-making regarding goal
TSH in managing hypothyroidism. One survey item explored the influence of patient
age, patient symptoms (such as racing heart tremors), patient preference, fatigue,
cardiac arrhythmias, heart disease, osteoporosis, fractures, multiple falls, and weight
on decision-making regarding goal TSH. Another survey item asked what TSH goal range
providers targeted in 8 clinical scenarios. The scenarios differed in patient sex
and age (30, 50, 65, or 85). For the clinical scenarios, TSH goal ranges were set
as 0.1 to 0.5, 0.6 to 1.5, 1.6 to 3.0, and 3.1 to 5.0 mIU/L.[30] Section III is a case-based survey discussing eight different fictional patients.
All patients were older females with a normal body mass index and experienced nonspecific
complaints resulting in fatigue. The females differed in age (70 vs. 85 years), vitality
status (vital vs. vulnerable disposition), and TSH (6 vs. 15 mU/L). All eight females
had a raised TSH, while free thyroxine was normal. Each case description was followed
by questions on the physicians' treatment strategy. They were asked (1) whether they
would start treatment and (2) what the starting dose of thyroxine would be.[31]
Statistical Analysis
We used STATA 15 software (Stata Corp LP, College Station, Texas, United States) for
data analysis. We summarized categorical data as percentages and continuous data as
mean ± standard deviation. We have no details of the nonresponders; hence, we did
not calculate the response rates. To assess the characteristics of physicians associated
with considering age when treating older adult patients with elevated TSH, we ran
a multivariate regression analysis that included the gender, region, age groups, years
of experience, specialty, grade, practice type, and proportion of patients > 65 years.
We used the chi-square test and Fisher's exact test to compare the response of physicians
(stratified by specialty) to the case scenarios. We used analysis of variance and
Bonferroni post hoc analysis to measure the differences in the mean levothyroxine
doses suggested by physicians (stratified by specialty) when treating older adult
patients. We considered p < 0.05 to be significant.
Results
Respondents' Characteristics
A convenience sample of physicians practicing in the MEA in relevant disciplines was
invited to take a Web-based survey of previously validated questionnaires to determine
the goal of TSH when treating hypothyroidism. A total of 286 surveys were received,
predominantly males, 180 (63.9%), mainly from the Middle East 194 (67.8%), and the
majority were senior physicians, 179 (62.6%), as shown in [Table 1]. Most physicians were endocrinologists, 117 (40.9%), followed by internal medicine,
88 (30.8%), and family medicine, 46 (16.1%).
Table 1
Demographic and professional characteristics of the study participants, including
the extent of their involvement in the general care of older adults and in diagnosing
and managing hypothyroidism in older adults
|
Sex (N = 286)
|
Male
|
180 (63.9%)
|
|
Female
|
106 (37.1%)
|
|
Region (N = 286)
|
Middle East
|
194 (67.8%)
|
|
Africa
|
58 (20.3%)
|
|
Southeast Asia
|
34 (11.9%)
|
|
Age groups
|
≤ 30 y
|
6 (2.1%)
|
|
31–40 y
|
65 (22.7%)
|
|
41–50 y
|
102 (35.7%)
|
|
51–60 y
|
75 (26.2%)
|
|
> 60 y
|
38 (13.3%)
|
|
Career stage (N = 286)
|
Junior
|
12 (4.5%)
|
|
Middle
|
78 (27.3%)
|
|
Senior
|
179 (62.6%)
|
|
Other
|
16 (5.6%)
|
|
Specialty (N = 286)
|
Endocrinology
|
117 (40.9%)
|
|
Internal medicine
|
88 (30.8%)
|
|
Family medicine
|
46 (16.1%)
|
|
Others
|
34 (12.2%)
|
|
Type of practice (N = 286)
|
Primary care
|
64 (22.4%)
|
|
Secondary care
|
65 (22.7%)
|
|
Tertiary care
|
148 (51.8%)
|
|
Others
|
9 (3.1%)
|
|
Years of experience (N = 286)
|
< 5 y
|
15 (5.3%)
|
|
5–10 y
|
43 (15.2%)
|
|
11–15 y
|
54 (19.2%)
|
|
16–20 y
|
50 (17.7%)
|
|
21–25 y
|
40 (14.2%)
|
|
> 25 y
|
80 (28.4%)
|
|
The percentage of patients in practice is 65 years or above (N = 276)
|
< 10
10–20
20–30
> 30
|
27%, 74
24%, 66
30%, 82
20%, 54
|
|
When did you last diagnose a patient over 65 with subclinical hypothyroidism? (N = 274)
|
< 1 week ago
< 1 month ago
< 1 year ago
< 3 years ago
> 3 years ago
|
19%, 51
34%, 92
26%, 72
7%, 20
14%, 39
|
|
When did you last start thyroxine treatment in a patient over 65 with subclinical
hypothyroidism? (N = 275)
|
< 1 week ago
< 1 month ago
< 1 year ago
< 3 years ago
> 3 years ago
Never
|
11%, 31
25%, 70
26%, 72
8%, 23
8%, 23
20%, 56
|
Respondent's Involvement in Hypothyroidism in Older Adults
The extent of the participant's involvement in the general care of older adults and
in diagnosing and managing hypothyroidism in this group is detailed in [Table 1]. About half of the respondents stated that > 20% of patients in their practice are
65 years or above; 24% reported having 10 to 20% of their patients in this age group.
When they were asked when they last diagnosed a patient over 65 with subclinical hypothyroidism,
19, 34, and 26% responded as less than 1 week, 1 month, and 1 year, respectively.
Starting thyroxine treatment in patients over 65 with subclinical hypothyroidism was
reported within the past week, past month, and past year by 11, 25, and 26%, respectively.
However, 20% never did it.
Impact of Age on Decision-Making Regarding TSH Target
[Fig. 1] shows the factors influencing the respondent's decision-making regarding the “goal
TSH range” when treating patients with thyroid hormone replacement therapy. Seventy-five
percent of physicians reported patient age as an important factor influencing decision-making
regarding goal TSH in treating hypothyroid patients. Patient age was scored fifth
to the presence of cardiac arrhythmia (80%), pregnancy (79%), heart disease (78%),
and patient symptoms (77%). Interest in conception (62%), fatigue (54%), weight (49%),
and osteoporosis (45%) were less commonly considered in decision-making. Fractures,
patient preference, and multiple falls were cited as relevant by even fewer respondents
(29, 19, and 18%, respectively). Only 3 (1%) physicians reported that none of the
above factors were important in their decision-making regarding goal TSH when treating
adults with hypothyroidism.
Fig. 1 The factors influencing the respondent's decision-making regarding the “goal thyroid-stimulating
hormone (TSH) range” when treating patients with thyroid hormone replacement therapy.
[Table 2] shows the univariate and multivariate analysis for considering patients' age (dependent
variable) in treating older adult patients with elevated TSH. Univariate analysis
showed that the decision to treat older adult patients varies according to the age
of the physicians (p = 0.037) and the specificity (p = 0.021). More endocrinologists consider age when treating older adult patients,
88 (75.2%), than internal medicine, 58 (65.9%), family physicians, 28 (60.9%), and
others, 17 (48.6%). However, multivariate analysis showed that only specialty impacts
the treatment decision in relation to age. Compared with endocrinologists, family
physicians and other specialties were less likely to consider age in their clinical
decisions, odds ratio (OR) 0.2 (95% confidence interval [CI] 0.1–0.7), p = 0.018, and OR 0.3 (95% CI 0.1–0.8), p = 0.013, respectively.
Table 2
Univariate and multivariable assessments of the characteristics of the physician linked
to the weight given to the patient's age when establishing the target TSH in hypothyroidism
|
Variables
|
Proportion of physicians
|
Univariate
p-value
|
Multivariate
odd ratio
|
Multivariate
p-value
|
|
Sex
|
Male
|
123 (68.3%)
|
0.536
|
Ref
|
|
|
Female
|
68 (64.8%)
|
1.1 (0.6–2.1)
|
0.702
|
|
Region
|
Africa
|
37 (63.85)
|
0.803
|
Ref
|
|
|
Middle East
|
132 (68.0%)
|
1.3 (0.6–2.9)
|
0.385
|
|
Southeast Asia
|
22 (64.7%)
|
0.8 (0.3–2.2)
|
0.676
|
|
Age group
|
< 20 y
|
0 (0.0%)
|
0.037
|
|
|
|
21–30 y
|
3 (75%)
|
Ref
|
|
|
31–40 y
|
39 (60.0%)
|
1.3 (0.8–21.1)
|
0.846
|
|
41–50 y
|
78 (76.4%)
|
4.7 (0.2–91)
|
0.310
|
|
51–60 y
|
45 (60%)
|
1.7 (0.1–37.0)
|
0.724
|
|
> 60 y
|
26 (68.4%)
|
2.0 (0.1–49.1)
|
0.651
|
|
Grades
|
Junior
|
11 (84.6%)
|
0.181
|
Ref
|
|
|
Middle
|
47 (60.2%)
|
0.15 (0.02–1.0)
|
0.053
|
|
Senior
|
120 (67.0%)
|
0.17 (0.02–1.2)
|
0.076
|
|
Other
|
13 (81.3%)
|
0.64 (0.07–6.3)
|
0.704
|
|
Specialty
|
Endocrinology
|
88 (75.2%)
|
0.021
|
Ref
|
|
|
Internal medicine
|
58 (65.9%)
|
0.7 (0.3–1.4)
|
0.323
|
|
Family medicine
|
28 (60.9%)
|
0.2 (0.1–0.7)
|
0.018
|
|
Others
|
17 (48.6%)
|
0.3 (0.1–0.8)
|
0.013
|
|
Practice
|
Primary care
|
41 (64.1%)
|
0.805
|
|
|
|
Secondary care
|
42 (64.6%)
|
0.6 (0.2–1.6)
|
0.300
|
|
Tertiary care
|
101 (68.2%)
|
0.7 (0.3–2.0)
|
0.541
|
|
Others
|
7 (77.8%)
|
2.3 (0.4–14.9)
|
0.376
|
|
Years of experience
|
< 5 y
|
30 (69.8%)
|
0.827
|
|
|
|
5–10 y
|
35 (64.8%)
|
Ref
|
|
|
11–15 y
|
33 (66.0%)
|
0.5 (0.2–1.5)
|
0.248
|
|
16–20 y
|
30 (75.0%)
|
0.5 (0.1–1.7)
|
0.261
|
|
21–25 y
|
50 (62.5%)
|
0.9 (0.2–4.0)
|
0.976
|
|
> 25 y
|
10 (66.7%)
|
0.7 (0.2–2.7)
|
0.677
|
|
Proportion of patients > 65 years
|
< 10 patients
|
44 (59.5%)
|
0.301
|
|
|
|
10–20 patients
|
45 (68.2%)
|
|
|
|
20–30 patients
|
60 (73.2%)
|
|
|
|
> 30 patients
|
34 (18.6%)
|
|
|
Abbreviation: TSH, thyroid-stimulating hormone.
[Table 3] shows the proportion of physicians who have decided to treat the patients in each
scenario for the group as a whole. Furthermore, [Table 4] shows that in all scenarios that included cases with subclinical hypothyroidism
(cases 1, 2, 5, and 8), endocrinologists and internal medicine physicians were less
likely to recommend treatment compared with family physicians and others (p < 0.05).
Table 3
The decision to treat or not to treat in a set of 8 fictional female cases of subclinical
hypothyroidism (T4 is within the reference range) by age, vitality status, and serum
TSH level and the starting dose per case scenario by the whole group of participants
|
Case
(no.)[a]
|
Age
(y)
|
Vitality
level
|
TSH level
|
Would you treat?[b]
|
Starting dose[c]
(mean ± SD)
(mcg)
|
|
Yes
|
No
|
|
1 (N: 246)
|
70
|
Vital
|
6 mU/L
|
20%, 50
|
80%, 196
|
31.6 ± 17.8
|
|
2 (N: 248)
|
85
|
Vulnerable
|
6 mU/L
|
15%, 37
|
85%, 211
|
34.6 ± 19.3
|
|
3 (N: 246)
|
70
|
Vital
|
15 mU/L
|
91%, 225
|
9%, 21
|
36.3 ± 18.0
|
|
4 (N: 243)
|
70
|
Vulnerable
|
15 mU/L
|
89%, 216
|
11%, 27
|
33.7 ± 19.1
|
|
5 (N: 246)
|
85
|
Vital.
|
6 mU/L
|
11%, 28
|
89%, 218
|
31.9 ± 16.2
|
|
6 (N: 244)
|
85
|
Vital
|
15 mU/L
|
82%, 200
|
18%, 44
|
30.9 ± 17.1
|
|
7 (N: 243)
|
85
|
Vulnerable
|
15 mU/L
|
79%, 191
|
21%, 52
|
29.8 ± 17.0
|
|
8 (N = 243)
|
70
|
Vulnerable
|
6 mU/L
|
17%, 41
|
83%, 202
|
33.5 ± 18.6
|
Abbreviations: SD, standard deviation; TSH, thyroid-stimulating hormone.
a Identification of the cases numbers are the same in [Tables 3] and [4]. Further clinical definitions are also provided in the questionnaire ([Supplementary Material S1], online only).
b Subgroup analysis of the decision to treat or not to treat by specialty is shown
in [Table 4].
c Subgroup analysis of the thyroxine starting dose per specialty is shown in [Table 4].
Table 4
The subgroup analysis was done by respondents' specialty of the proportion of physicians
deciding to start treatment (I) and the mean starting dose (II) in each case scenario
|
Case[a]
|
Endocrinology
|
Internal medicine
|
Family medicine
|
Others
|
p-Value
|
|
I. The proportion of physicians deciding to start treatment, N (%):
|
|
1 (N: 246)
|
18 (17.1)
|
10 (13.0)
|
13 (35.1)
|
9 (33.3)
|
0.011
|
|
2 (N: 248)
|
9 (8.6)
|
7 (9.0)
|
14 (37.8)
|
7 (25.0)
|
< 0.001
|
|
3 (N: 246)
|
102 (98.1)
|
68 (87.2)
|
33 (91.7)
|
22 (78.6)
|
0.003
|
|
4 (N: 243)
|
95 (89.5)
|
69 (90.8)
|
31 (86.1)
|
22 (84.6)
|
0.781
|
|
5 (N: 246)
|
9 (8.6)
|
2 (2.6)
|
9 (25.0)
|
8 (29.6)
|
< 0.001
|
|
6 (N: 244)
|
88 (85.4)
|
63 (80.8)
|
28 (77.8)
|
21 (77.8)
|
0.645
|
|
7 (N: 243)
|
79 (76.0)
|
62 (80.5)
|
30 (83.3)
|
20 (76.9)
|
0.772
|
|
8 (N = 243)
|
14 (13.4)
|
9 (11.7)
|
12 (34.3)
|
6 (22.2)
|
0.015
|
|
II. The mean start dose of levothyroxine therapy (mcg/day)
|
|
3 (N: 246)
|
34.5 ± 14.2
|
33.0 ± 14.7
|
45.2 ± 27.7
|
42.0 ± 20.6
|
0.004
|
|
4 (N: 243)
|
28.1 ± 11.1
|
32.3 ± 16.2
|
46.5 ± 29.7
|
44.0 ± 24.5
|
< 0.001
|
|
6 (N: 244)
|
27.0 ± 11.0
|
27.8 ± 10.0
|
45.7 ± 30.2
|
36.6 ± 20.0
|
< 0.001
|
|
7 (N: 243)
|
24.8 ± 9.6
|
27.3 ± 11.8
|
44.0 ± 27.6
|
37.5 ± 21
|
< 0.001
|
a Identification of the case numbers are the same in [Tables 3] and [4]. Further clinical definitions are also provided in the questionnaire ([Supplementary Material S1], online only).
The recommended start dosage of levothyroxine in all cases of hypothyroidism is shown
for the whole group in [Table 3] and by specialty in [Table 4]. Family and other physicians recommend higher starting doses than endocrinologists
and internal medicine physicians. Post hoc analysis showed that family physicians
recommend significantly higher doses of levothyroxine in all case scenarios than endocrinologists
and internal medicine physicians.
Factors Determining the Goal TSH
[Fig. 2] shows the distribution of the respondents' answers to clinical scenarios differing
in patient age and sex regarding goal TSH range in hypothyroid patients. It is shown
that around 90% of respondents aim for a TSH ≤ 3.0 mIU/L in a 30-year-old patient
regardless of sex. This proportion decreases with age such that 15% of respondents
targeted a TSH of 1.6 to 3.0 mIU/L in octogenarians, but 78% targeted a TSH of 3.1
to 5.0 mIU/L in this group. Regardless of sex, physician-reported TSH goal ranges
(0.1–0.5, 0.6–1.5, 1.6–3.0, and 3.1–5.0 mIU/L) increased directly to patient age.
Fig. 2 The percent participants' chosen goal thyroid-stimulating hormone (TSH) range mU/L
(on the X-axis) for each of the age groups (30, 50, 65, and 85) and sex (females above and
males below) (N = 252).
Impact of Age, Vitality, and TSH on the Management of Subclinical Hypothyroidism
Overall, respondents were less inclined to start treatment in 85-year-olds than in
70-year-old females with TSH of 6 mU/L (20% vs. 11%). Females with a TSH of 15 mU/L
were more likely to get treated than those with a 6 mU/L TSH. Vital persons are more
likely to be treated with thyroxine than vulnerable persons for the same TSH levels.
When indicated, the dose does not seem different for the various groups other than
a tendency for a lower dose in those vulnerable persons over 85 years ([Table 3]). Patterns were similar when comparing the TSH goals in female and male patients,
which differed by age ([Fig. 2]).
Discussion
In the present survey, we found a considerable variation in physicians' treatment
strategies for older people with subclinical hypothyroidism. We reported that 75%
of respondents considered the patient's age as an important factor in their decision-making
when determining TSH goals in hypothyroid patients. However, when presented with more
granular clinical scenarios that incorporated patient sex and age, physicians' reported
TSH goals shifted with age but not sex. Even though in our survey we found that a
majority (78%) aim for a TSH of 3.1 to 5.0 mIU/L in octogenarians, those highest at
risk for adverse outcomes with excess thyroid hormone replacement, over 90% of respondents
aim for a TSH > 1.5 mIU/L in the 85-year-old category, which may suggest an attempt
to avoid overtreatment in this age group. However, this target still would not adequately
avoid overtreatment. A similar pattern is not seen in the 30-year-old group, the cohort
at lowest risk for adverse effects from thyroid hormone overtreatment. Our findings
indicate that endocrinologists and internal medicine specialists consider age more
when making decisions than family physicians and other specialties. We consistently
observed that family physicians are more inclined to treat individuals over 65 years
with subclinical hypothyroidism as opposed to other specialties. Finally, we established
that endocrinologists and internal medicine physicians recommend lower starting doses
of levothyroxine for people with hypothyroidism who are above the age of 65 years.
Over a decade ago, a survey of primary care physicians and thyroid specialists regarding
the management of hypothyroidism showed that only 30% of primary care physicians and
19% of thyroidologists aimed for a higher TSH range (2.0–5.0 mIU/L) for a 73-year-old
patient with hypothyroidism.[32] Another recent study that exclusively surveyed clinical endocrinologists showed
that 29.5% would target a TSH range of 3.0 to 3.9 mIU/L and 22.5% would target a TSH
range of 4.0 to 4.9 mIU/L in an octogenarian.[30] Our study concurs with previous studies that showed no consensus among physicians
regarding TSH goals depending on patient age.[30]
[31]
There is reasonably convincing evidence that TSH increases with age,[2]
[3]
[4]
[5] and lowering this with thyroid hormone therapy may increase the risk of overtreatment.[6]
[7]
[8] Previous population studies have shown that increasing patient age is associated
with shifts in the TSH distribution curve to higher concentrations, presumably due
to changes in TSH bioactivity, changes in negative feedback relationships between
TSH and FT4, or influences from medications.[2]
[17]
[18] Additionally, several other studies have demonstrated an association between longevity
and higher TSH levels in several patient populations, suggesting that mildly elevated
TSH levels may offer potential benefits to certain older adult patients.[3]
[4] Several guidelines recognize that the normal TSH reference range changes with age[19]
[20] but do not recommend targeting specific TSH goals based on patient age. Perhaps
age-specific TSH ranges should be implemented to avoid overtreatment in vulnerable
populations, such as older adults. A general population study demonstrated that only
60% of adults taking thyroid hormone have TSH levels in the euthyroid range, with
up to 20% being overtreated.[8] In a cohort of men and women aged 65 years and older enrolled in the Cardiovascular
Health Study, 41% of thyroid hormone users were found to be overtreated.[9] The deleterious effects of excess thyroid hormone on the cardiac system, including
cardiac arrhythmias and increased cardiovascular mortality, and skeletal systems,
including osteoporosis, have been described in several cohorts.[10]
[11]
[12]
[13]
[14]
[15]
[16] In addition, higher TSH and lower FT4 concentrations within the euthyroid range
in older patients (age ≥ 65 years) not taking thyroid hormone medications are associated
with a lower risk of multiple adverse events in older adults, including mortality.[11] In this current study, in addition to evaluating the role of age, we investigated
other reasons for modifying TSH goals. Interestingly, cardiac disease, pregnancy,
and symptoms (such as tremors and racing heartbeat) were considered the most important
factors influencing goal TSH for patients with hypothyroidism. This may be appropriate
as symptoms may indicate overtreatment with thyroid hormone replacement, and dose
adjustments may be necessary. Interestingly, over half of the treating physicians
cited fatigue as an important factor influencing decision-making, contradicting current
guidelines, which do not recommend adjusting the TSH goal in hypothyroid patients
according to symptoms such as fatigue.[17]
Furthermore, despite the well-documented adverse effects of a low TSH on the cardiac
and skeletal systems, physicians were more likely to consider cardiac arrhythmias
and heart disease rather than osteoporosis, fractures, and falls in their decision-making
process.[10]
[11]
[12]
[13]
[14]
[15]
[16] This discrepancy underscores the need for more comprehensive physician education
to emphasize the skeletal risks associated with thyroid hormone overtreatment and
cardiac risks, particularly in older adults.
In the section on subclinical hypothyroidism, a large variation in physicians' treatment
strategies for older people with subclinical hypothyroidism was found by patient characteristics.
Similar to the international study, the results reflect the lack of global consensus
and the large variation in guidelines on whether or not to treat subclinical hypothyroidism,
especially in older people.[21]
[32]
[33] Although guidelines often offer specific recommendations for the treatment of older
patients with overt hypothyroidism, they usually lack explicit guidance for managing
older patients with subclinical hypothyroidism.[34]
[35]
The main strength of this study is its pioneering nature, which is the first of its
kind to address this issue across a large geographical area of developing countries.
We included various specialties in managing hypothyroidism in older adults and employed
a comprehensive set of independent variables in a succinct online survey with diverse
questions. However, notable limitations exist. The convenience sampling method used
precludes the calculation of a reliable response rate, and there is a risk of nonresponse
bias inherent in survey studies. While the survey encompassed an extensive array of
factors influencing physician decision-making in managing hypothyroidism, other factors
could have been included.
Moreover, discrepancies may exist between providers' responses to survey questions
and their actual treatment behaviors. Furthermore, the focus solely on physicians
overlooks the involvement of patients in the decision-making process. Additionally,
physicians were not presented with cases involving differences in symptoms or various
procedure options, such as watchful waiting or repeated thyroid function assessments.
These limitations underscore the need for caution when interpreting the findings and
suggest avenues for future research to address these gaps in knowledge and methodology.
Conclusion
This study has implications for both patients and physicians. Failure to recognize
the relationship between age and thyroid hormone requirements may lead to overtreatment,
especially in older adults. Three-quarters of all physicians consider patients' age
an important factor when determining TSH goals. When presented with clinical scenarios
differing by patient sex and age, 90% of respondents reported aiming for a TSH > 1.5
mIU/L in octogenarians. This may indicate an attempt to avoid overtreatment in this
specific age group. The observed physicians' treatment strategies for older people
with subclinical hypothyroidism vary largely with patients' characteristics. The observed
significant interphysician variations reflect the need for more evidence-based guidelines
and increased awareness of treating physicians across specialties.
