Prevalence and predictors of adequate treatment of overt hypothyroidism - a population-based study

Abstract

The aim of this study is to evaluate the adequacy of treatment, and to identify factors influencing treatment of hypothyroidism. Patients newly diagnosed with overt hypothyroidism (n=345) were identified via a register linked to a laboratory database. In selected periods with staff available, 165 patients were invited, and 113 (68.5 %) accepted participating in a comprehensive program including blood tests and completion of questionnaires. We performed a longitudinal follow-up on thyroid function tests 10 years after the diagnosis. Time to reach a serum TSH level of 0.2-10 mU/L (termed as clinically acceptable) and biochemical normalization (TSH: 0.2-5.0 mU/L), respectively, were analyzed using Kaplan Meier survival analysis. Predictors for longer duration to reach the normal TSH range were identified using cox proportional hazards regression. Only 67.7 % of the patients were in the euthyroid range on the long term after diagnosis of overt hypothyroidism (2 years: 59.4 %; 10 years: 67.7 %). Median time to the first normal TSH was 8.9 months (95 % CI: 7.6-10.2 months). The factors associated with longer duration until normalization of TSH after multivariate analysis were age (HR 0.79 per 10 years; 95 % CI: 0.66-0.94; P = <0.01), smoking (HR 0.47; 95 % CI: 0.26-0.83; P = <0.01), serum TSH at diagnosis (HR 0.96 per 10 mU/L; 95 % CI: 0.93-0.99; P = 0.02) and BMI (HR 0.96 per kg/m²; 95 % CI: 0.91-0.99; P = 0.03). A considerable number of hypothyroid patients remained inadequately treated. When treating hypothyroid patients, special attention should be addressed to those patients who never or lately obtain euthyroid status.

Keywords: hypothyroidism; overt hypothyroidism; population-based study; treatment.

Table 1. Baseline characteristics of participants in the investigational program and the non-participants

Table 2. Univariate analysis showing median months to first normal TSH among subgroups within various clinical and demographic variables

Table 3. Multivariate analysis with cox proportional hazards regression, showing hazard ratios (HRs) with 95 % confidence intervals (CI) for first TSH within normal range according to clinical and demographic factors

Figure 1. Flowchart depicting the origin and selection of hypothyroid cases (n=345) and participants in investigational program (n=113) during the study

Figure 2. Distribution of thyroid status over time among patients who had a TFT performed within each of the time ranges

Figure 3. Kaplan Meier curves for the fraction of patients who had a TFT performed within the clinically acceptable and normal TSH range, respectively. Censoring is indicated by the black dot tic mark. For visual purpose the figure does not show the entire follow-up period (11 years). Therefore, few cases are not shown in the figure due to very late occurring normalization. 1 case was left out on the clinically acceptable curve (time 121.9 months). 3 cases were left out on the normal TSH curve (time 117.9, 121.9 and 122.1 months). On the clinically acceptable curve, only the 272 patients who had TSH ≥10 mU/L at diagnosis are included.