Abstract
Objective: To evaluate outcomes associated with adherence to levothyroxine (LT4) in the US adult hypothyroidism population.
Methods: We used data from Truven’s MarketScan databases from 1 July 2011 through 31 December 2015. Patients aged 18 or older were diagnosed with hypothyroidism (confirmed at least twice) and prescribed LT4. Patients were excluded if they did not have continuous insurance coverage or if they received a diagnosis of thyroid cancer or pregnancy during the study period. Multivariable analyses on a matched cohort of adherent and nonadherent patients examined the relationships among patient outcomes and adherence, defined as the proportion of days covered ≥80%. Outcomes included all-cause and hypothyroidism-related medical costs and resource utilization and comorbid diagnoses measured over the 1 year post-period following the first prescription for LT4. The analyses controlled for patient age, sex, region of residence, type of insurance coverage, diagnosing physician and pre-period general health status as proxied by the Charlson Comorbidity Index.
Results: Prior to matching, there were 168,457 patients identified as adherent and 198,443 patients identified as nonadherent. The matched cohort consisted of 318,628 individuals, with equal numbers of adherent and nonadherent patients (n = 159,314). Patients who were adherent used significantly fewer resources and had significantly lower all-cause ($14,136 vs. $14,926; p < .0001) and hypothyroidism-related ($1672 vs. $1709; p < .0001) total costs, although the costs of drugs were higher in the adherent group. Furthermore, adherent patients, compared to nonadherent patients, were significantly less likely to be diagnosed with comorbid Addison’s disease, bipolar disorder, chronic kidney disease, depression, migraine, obesity, type 1 diabetes or type 2 diabetes during the follow-up period.
Conclusions: Compared to nonadherence, adherence to LT4 among patients with hypothyroidism was associated with a significant reduction in all-cause and hypothyroidism-related costs and resource utilization as well as significantly lower rates of many comorbid diagnoses.
Introduction
Low thyroid function (hypothyroidism) affects between 4.6% and 9.5% of the US populationCitation1,Citation2. While hypothyroidism may influence multiple organ systems and can have substantial impact on patient well-being, symptoms vary and may be mistakenly attributed to other diseasesCitation3,Citation4. Some of the more common symptoms include fatigue, constipation, weight gain, carpal tunnel syndrome, dry skin and goiterCitation5. Uncontrolled hypothyroidism has also been linked to more serious comorbidities, including cognitive dysfunctionCitation6,Citation7, mood disordersCitation8,Citation9 and cardiovascular diseaseCitation10.
The standard treatment for hypothyroidism is thyroid hormone replacement therapy with levothyroxine (LT4)Citation3. LT4 must be taken consistently to normalize thyroid hormone levelsCitation11. However, one study has indicated that 31.6% of hypothyroid patients fail to take their LT4 medication regularly during the first year post-initiationCitation12. That study was published nearly a decade ago. No previous research has reported the economic costs of nonadherence to LT4.
We therefore began our research with two main objectives: first, to evaluate adherence to LT4 therapy using more recent data and, second, to quantify costs and comorbidities associated with nonadherence. Through these analyses, we described the demographic, clinical and treatment characteristics of patients who initiated treatment with LT4. We also examined the differences between those who were adherent to LT4 therapy and those who were nonadherent.
Methods
The study data came from two sources: the Truven Health Analytics MarketScan Commercial Claims and Encounters (CCAE) database provided healthcare information for commercially insured individuals who were covered by fully or partially capitated, fee for service plans and the Medicare Supplemental and Coordination of Benefits (MDCR) database supplied the healthcare experiences of retirees with Medicare supplemental insurance paid for by employersCitation13. Both databases contained detailed information on healthcare costs, resource use, and outcomes for inpatient and outpatient services. We examined records dated from 1 July 2011 through 31 December 2015. All the information was fully de-identified and compliant with the Health Insurance Portability and Accountability Act (HIPAA).
To be included in the study, we required that patients have two separate claims for hypothyroidism (identified by ICD-9-CM of 243.xx or 244.xx, except 244.2x and 244.3x) between 1 January 2012 and 31 December 2014 and at least one prescription for levothyroxine (LT4) filled over that same time horizon. The date on which each patient first filled an LT4 prescription marked the index date. We furthermore required all patients to have had continuous insurance coverage from the start of the 6 month pre-period through the end of the 1 year post-period. Individuals were excluded from the analysis if they were younger than age 18 as of the index date or if they were diagnosed with thyroid cancer (ICD-9-CM of 193.xx) or pregnancy (ICD-9-CM of 630.xx–679.xx, V22.xx, V23.xx, V30.xx–V39.xx or V72.42) at any time from the start of the pre-period through the end of the post-period. The xx signifies that there can be any characters after the first 3 digits and it will be included (for example 630.xx means we would include all diagnoses between 630.00 and 630.99). These criteria resulted in a sample of 366,918 individuals. illustrates how each of the inclusion and exclusion criteria affected the sample size.
Figure 1. Inclusion/exclusion criteria and sample size.
Given the cohort described above, we categorized patients as adherent or nonadherent using the proportion of days covered (PDC). The PDC is used by both the Pharmacy Quality AllianceCitation14 and the Centers for Medicare and Medicaid servicesCitation15 and is defined as the percentage of days an individual received levothyroxine in the post-period. Consistent with previous researchCitation16–18, we defined patients as adherent when they took LT4 at least 80% of the days in the post-period, as indicated by their insurance claims for prescriptions filled. Nonadherent patients were matched 1:1 without replacement to adherent patients based upon age, sex, region of residence and type of insurance coverage.
The study focused on the relationship between adherence to LT4 therapy and patient outcomes, including costs, resource use and comorbidities. The major cost categories were all-cause and hypothyroidism-related. The cost subcategories were inpatient, outpatient, emergency room (ER), laboratory and drugs. To measure costs, we assessed gross payments to a provider for a service and converted all amounts to 2015 prices using the medical component of the consumer price indexCitation19. We assessed resource use by determining the probability of a hospitalization or an ER visit; the hospital length of stay (LOS); and the number of hospitalizations, ER visits and outpatient visits. We analyzed all-cause and hypothyroidism-related resource use. Finally, we examined the probability of being diagnosed with certain comorbidities, including Addison’s disease, bipolar disorder, celiac disease, chronic kidney disease, depression, dyslipidemia, heart failure, hypertension, migraine, myasthenia gravis, obesity, pernicious anemia, rheumatoid arthritis, schizophrenia, systemic lupus erythematosus, type 1 diabetes and type 2 diabetes. We included comorbidities linked to hypothyroidism either by previous researchCitation8,Citation20–22 or by the American Association of Clinical Endocrinologists and the American Thyroid Association (AACE/ATA) clinical practice guidelines for hypothyroidism in adultsCitation23.
To investigate the relationship between the outcomes described above and patient adherence, we applied multivariable analyses to the matched cohort of adherent and nonadherent patients. In all multivariable analyses, we controlled the individual patient’s characteristics (age, sex, region of residence and type of insurance coverage), pre-period general health status as proxied by the Charlson Comorbidity Index (CCI)Citation24 and whether the patient visited an endocrinologist in the pre-period. To study outpatient, drug and total costs, we used general linear models with a gamma distribution and log link. To examine inpatient, ER and laboratory costs, we employed two-part models: the first part captured the probability of the event of interest, while the second part estimated the costs of the users of the service. We used negative binomial regressions to determine hospital LOS and the number of hospitalizations, ER visits and outpatient visits, and logistic regressions to examine the probability of a hospitalization, an ER visit or a diagnosis of a comorbidity.
After conducting the multivariable analyses, we examined differences in the estimated outcomes between adherent and nonadherent patients using t-tests or odds ratios (ORs) and 95% confidence intervals (CIs). We conducted all analyses using SAS version 9.4 and considered p values <.05, a priori, to be statistically significant.
Results
presents descriptive statistics for patients both before and after matching, based upon adherence status. Prior to matching, 168,475 of the 366,918 patients (45.92%) were found to be adherent over the 1 year post-period, and these patients were significantly older (54.64 years vs. 51.74 years; p < .0001) and more likely to be male (26.10% vs. 24.02%; p < .0001). There were also significant regional (p < .0001) and insurance-type (p < .0001) differences, with adherent patients more likely to reside in the Northeast (16.43% vs. 13.00%) and less likely to reside in the South (37.73% vs. 40.66%) or to have preferred provider organization insurance (56.87% vs. 58.44%). Matching resulted in a successful match rate of 94.56%. By construction, matching removed differences in patient characteristics. However, post-matching, adherent patients were significantly more likely than nonadherent patients to have been diagnosed by an endocrinologist (6.73% vs. 5.91%; p < .0001), and they had a significantly lower CCI score (0.60 vs. 0.62; p < .0001).
Table 1. Descriptive statistics pre- and post-matching by adherence status.
and focus on the association between adherence to LT4 therapy and all-cause and hypothyroidism-related costs. As shown in , adherence was associated with significantly lower all-cause outpatient ($6997 vs. $7269; p < .0001), inpatient ($2655 vs. $3470; p < .0001), ER ($568 vs. $753; p < .0001), laboratory ($89 vs. $116; p < .0001) and total ($14,136 vs. $14,926; p < .0001) costs, despite being associated with significantly higher drug costs ($3321 vs. $2807; p < .0001). shows that these same general results are found when examining hypothyroidism-related costs. Specifically, despite higher drug acquisition costs associated with adherence ($130 vs. $82; p < .0001), adherent patients had significantly lower hypothyroidism-related total costs compared to nonadherent patients ($1672 vs. $1709; p < .0001). These significantly lower hypothyroidism-related costs are consistent with the finding that outpatient ($777 vs. $786; p < .0001), inpatient ($770 vs. $876; p < .0001), ER ($130 vs. $158; p < .0001) and laboratory ($19 vs. $23; p < .0001) costs were all significantly lower for patients who were adherent to LT4 therapy compared to nonadherent patients.
Figure 2. All-cause direct medical costs: by adherence threshold.
Figure 3. Hypothyroidism-related costs: by adherence threshold.
examines differences in all-cause and hypothyroidism-related resource utilization between adherent and nonadherent patients. Consistent with the finding that inpatient and ER costs were lower for adherent patients, all-cause and hypothyroidism-related hospital and ER resource use were significantly lower for these patients as well. Compared to nonadherent patients, adherent patients were significantly less likely to have an all-cause hospitalization (9.15% vs. 10.88%; p < .0001) or ER visit (19.97% vs. 23.83%; p < .0001), and they were also less likely to have a hypothyroidism-related hospitalization (4.43% vs. %.15%; p < .0001) or ER visit (5.26% vs. 6.10%; p < .0001). In addition, adherent patients had significantly fewer all-cause or hypothyroidism-related ER visits or hospitalizations. Furthermore, among all the patients who were hospitalized, those who were adherent had significantly lower all-cause hospital LOS (5.84 days vs. 7.28 days; p < .0001) and hypothyroidism-related hospital LOS (4.96 days vs. 5.63 days; p < .0001). In contrast, adherence with LT4 was associated with significantly more all-cause (17.31 vs. 17.01; p < .0001) and hypothyroidism-related (2.65 vs. 2.54; p < .0001) outpatient visits, although the costs associated with such visits were significantly lower for adherent patients.
Table 2. All-cause and hypothyroidism-related resource utilization by adherence status.
illustrates the association between the likelihood of being diagnosed with a comorbidity of interest in the post-period and being adherent to LT4 therapy. Results indicate that, compared to nonadherent patients, adherent patients were significantly less likely to be diagnosed with comorbid Addison’s disease, bipolar disorder, depression, heart failure, migraine, obesity, type 1 diabetes or type 2 diabetes. For example, adherent patients were 20% less likely to be diagnosed with Addison’s disease (odds ratio [OR] = 0.801; 95% confidence interval [CI] 0.722–0.888), 7% less likely to be diagnosed with bipolar disorder (OR = 0.929; 95% CI 0.886–0.875), and 11% less likely to be diagnosed with type 1 diabetes (OR = 0.888; 95% CI 0.847–0.931). In contrast, adherent patients were significantly more likely to be diagnosed with comorbid dyslipidemia (OR = 1.046; 94% CI 1.031–1.062). The adherent and nonadherent patients had the same statistical likelihood of being diagnosed with comorbid celiac disease, chronic kidney disease, hypertension, myasthenia gravis, pernicious anemia, rheumatoid arthritis, schizophrenia or systemic lupus erythematosus. However, less than 1% of the population had a diagnosis of celiac disease, myasthenia gravis, pernicious anemia, schizophrenia or systemic lupus erythematosus, making it difficult to identify clinically meaningful differences.
Figure 4. Association between comorbidities and patient adherence: odds ratios (ORs) and 95% confidence intervals (CIs).
As a test of the sensitivity of the results, we conducted separate analyses for individuals aged <65 years and those ≥65 years at the index date. As presented in Supplementary Tables 1 and 2 and Supplementary Figures 1 and 2, these alternate results are generally consistent with the primary analyses. However, as expected, the cohort aged ≥65 generally had more comorbid diagnoses, used more resources and had higher costs compared to the younger cohort. In addition, two significant differences emerged when we stratified patients by age. First, adherent patients younger than age 65 had significantly higher hypothyroidism-related, all-cause and outpatient costs compared to nonadherent patients. We believe that the larger costs likely represent increased outpatient monitoring, consistent with better management of hypothyroidism. In addition, there were fewer significant differences in comorbidities between adherent and nonadherent patients age 65 or older. Specifically, among patients age 65 or older, adherent and nonadherent patients had the same likelihood of being diagnosed with Addison’s disease, bipolar disorder, obesity or type 2 diabetes. In contrast, when all ages were combined, adherent patients were significantly less likely to be diagnosed with these conditions.
Discussion
Consistent with adherence literature from other disease statesCitation25–27, the present analyses provide real-world evidence that patients with hypothyroidism who adhere to LT4 therapy have substantially better outcomes relative to their nonadherent counterparts. The results also suggest that a majority of US patients with hypothyroidism are nonadherent to LT4 therapy. The 54.08% rate of nonadherence in the present study is higher than the 31.6% reported in a 2008 US study that used claims data from a decade ago (2001–2004)Citation12. This higher rate of nonadherence is consistent with research which has found an increase in the percentage of personal healthcare expenditures paid for prescription drugs as well as an increase in the percentage of individuals using five or more prescription drugs over this periodCitation28.
Adherent patients, compared to nonadherent patients, were found to have significantly lower total medical costs, despite the higher drug costs associated with adherence. In this study, adherent patients had annual total costs that were $790 lower than nonadherent patients and non-drug costs that were $1304 lower. Given our matched sample of 159,314 nonadherent individuals, this suggests the potential of annual total cost savings of $125.8 million and non-drug cost savings of $207.7 million in this population if all patients were adherent. Given that hypothyroidism is a chronic condition often treated for decades and the condition affects between 4.6% and 9.5% of the US populationCitation1,Citation2, there appears to be a large economic impact associated with nonadherence.
In addition to using more medication, the adherent patients in the present study used more outpatient services but less acute care relative to their nonadherent counterparts. This pattern of resource use has important clinical and economic implications. For instance, both outpatient care and regular drug use are planned, maintenance activities which indicate patient engagement. Previous research has shown that outpatient care is associated with better disease controlCitation29. In contrast, acute care is episodic and potentially traumatic in nature. Previous research has shown that acute care is associated with a number of adverse consequences, including loss of productivityCitation30, healthcare-associated infectionsCitation31,Citation32 and medical errorsCitation33,Citation34. Moreover, acute care is the largest component of direct medical spendingCitation35, in concert with the present finding that the nonadherent patients had substantially higher total costs.
In addition to higher total costs, the nonadherent patients in this study also had more comorbid illness relative to the adherent cohort, in agreement with previous literature reporting that there are “substantially greater, near exponential, increases in healthcare costs and resource utilization when additional chronic comorbid conditions are present”Citation36. In particular, the nonadherent cohort had higher costs for outpatient services and laboratory work, possibly because this cohort used such services more often for the management of severe illness than for routine follow-up. For instance, in addition to having more hypothyroidism-related resource use and costs, the nonadherent patients were more likely to have a number of burdensome comorbidities, including Addison’s disease, bipolar disorder, depression, heart failure, migraines and obesity. The relationship between patient comorbidities and adherence is consistent with research which has found that polypharmacy and regimen complexity adversely affect patient adherenceCitation37,Citation38.
While all of these comorbid conditions have known associations with hypothyroidismCitation10,Citation22,Citation39–42, several have been shown to have particularly complex relationships with LT4 treatment, which may in turn have affected adherence. For instance, lithium treatment for bipolar disorder can reduce thyroid functionCitation43 while reduced thyroid function can influence the course of depression and bipolar disorderCitation39. Furthermore, treatment with thyroxine has been found to have antidepressant and prophylactic effects for patients with unipolar and bipolar depressionCitation44, although there is stronger evidence for use of triiodothyronine as an antidepressant augmentation therapyCitation45. Patients with heart failure were found to be less likely to be adherent, possibly due to fears of LT4 overtreatment, which has been shown to increase the risk of irregular or rapid heartbeat, particularly among older patientsCitation46,Citation47. However, untreated hypothyroidism has been shown to increase the risk of heart failureCitation48,Citation49 as well as coronary heart disease (CHD) and CHD-related deathCitation10. In this study, adherent patients were found to be more likely to have comorbid dyslipidemia, despite the fact that LT4 treatment usually corrects dyslipidemiaCitation50. However, in cases where LT4 therapy does not correct dyslipidemia, other factors, such as genetics, age or diet, may be involvedCitation50.
The findings of this study must be interpreted in the context of the limitations. First, the commercially insured patients in this study may not represent most Americans with hypothyroidism. For example, given that these patients are relatively well insured, they may do a better job of maintaining their health relative to patients who are less well insured or uninsured. As a result, they may be in better general health and use fewer medical resources compared to other groups of US patients. Second, we were unable to control for undocumented factors, such as diet, severity of illness, over-the-counter medication use, race, smoking status, income level and education status that may affect outcomes. Furthermore, the analyses do not control for the use of concomitant antithyroid drugs or the use of other drugs, such as lithium, that may affect thyroid function. Third, the insurance claims revealed prescriptions filled but not whether or how the LT4 was taken. In addition, the database contained no information on laboratory test results. As a result, we were unable to examine any associations between adherence to LT4 and thyroid function. Furthermore, it should be noted that this analysis does not examine interactions between comorbid conditions and patient adherence. For example, hypothyroidism has been shown to be associated with dyslipidemia and patients with dyslipidemia have an increased cardiovascular riskCitation51. Finally, our analyses focused on statistical significance but was not able to address whether differences in outcomes represented minimally important clinical differences.
Conclusion
In this study of patients with hypothyroidism, individuals who were adherent to levothyroxine had substantially better outcomes relative to those who were nonadherent. In particular, the adherent patients had a reduced likelihood of being admitted to a hospital or an ER, a lower mean number of hospitalizations or ER visits, and a shorter mean length of hospital stay. Although the cost of the drugs was slightly higher for the adherent patients, they nevertheless had lower costs in every other category, including total, hospital, ER, outpatient and laboratory costs. These outcomes demonstrate that improving LT4 adherence has the potential to substantially benefit patients and payers.
Transparency
Declaration of funding
Financial support for this study was provided by AbbVie Inc. AbbVie participated in the study design, data collection, analysis, interpretation of results, writing, reviewing and approving of publication.
Author contributions: M.J.L. had primary responsibility for conducting the statistical analyses of the data. All authors were responsible for the study concept and design, interpretation of the data, drafting of manuscript, and revising of the manuscript.
Declaration of financial/other relationships
Z.H. has disclosed that he is a former employee and stockholder of AbbVie Inc. R.E. has disclosed that he is an employee and stockholder of AbbVie Inc. M.J.L. has disclosed that she is an employee of HealthMetrics, which received funding from AbbVie for the conduct of this study. V.V.G. has disclosed thathe is a research consultant for AbbVie.
JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
Supplemental Material
Download MS Word (116.1 KB)Acknowledgements
The authors would like to acknowledge Patricia Platt and Michael Treglia of HealthMetrics Outcomes Research for assistance during manuscript preparation, for which AbbVie provided compensation.
References
- Canaris GJ, Manowitz NR, Mayor G, et al. The Colorado thyroid disease prevalence study. Arch Intern Med 2000;160:526-34
- Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2002;87:489-99
- Jonklaas J, Bianco AC, Bauer AJ, et al. Guidelines for the treatment of hypothyroidism: prepared by the American Thyroid Association Task Force on Thyroid Hormone Replacement. Thyroid 2014;24:1670-751
- Vanderpump MP, Tunbridge WM. Epidemiology and prevention of clinical and subclinical hypothyroidism. Thyroid 2002;12:839-47
- El-Shafie KT. Clinical presentation of hypothyroidism. J Family Community Med 2003;10:55-8
- Miller KJ, Parsons TD, Whybrow PC, et al. Verbal memory retrieval deficits associated with untreated hypothyroidism. J Neuropsychiatry Clin Neurosci 2007;19:132-6
- Hogervorst E, Huppert F, Matthews FE, et al. Thyroid function and cognitive decline in the MRC Cognitive Function and Ageing Study. Psychoneuroendocrinology 2008;33:1013-22
- Radhakrishnan R, Calvin S, Singh JK, et al. Thyroid dysfunction in major psychiatric disorders in a hospital based sample. Indian J Med Res 2013;138:888-93
- Khemka D, Ali JA, Koch CA. Primary hypothyroidism associated with acute mania: case series and literature review. Exp Clin Endocrinol Diabetes 2011;119:513-17
- Rodondi N, den Elzen WP, Bauer DC, et al. Subclinical hypothyroidism and the risk of coronary heart disease and mortality. JAMA 2010;304:1365-74
- Scavone C, Sportiello L, Cimmaruta D, et al. Medication adherence and the use of new pharmaceutical formulations: the case of levothyroxine. Minerva Endocrinol 2016;41:279-89
- Briesacher BA, Andrade SE, Fouayzi H, et al. Comparison of drug adherence rates among patients with seven different medical conditions. Pharmacotherapy 2008;28:437-43
- Hansen LG, Chang S. Health research data for the real world: the MarketScan databases. Truven Health Analytics, 2011. Available at: http://truvenhealth.com/portals/0/assets/PH_11238_0612_TEMP_MarketScan_WP_FINAL.pdf [Last accessed 18 November 2017]
- Pharmacy Quality Alliance. PQA Medication Quality Measures in the Health Insurance Marketplace. Alexandria (VA): PQA, 2015. Available at: http://pqaalliance.org/measures/qrs.asp [Last accessed 18 November 2017]
- Centers for Medicare and Medicaid Services. 2017 Medicare–Medicaid plan performance data technical notes. Quality Rating System Measure Technical Specifications. Washington (DC): CMS, 2017. Available at: https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medi-caid-Coordination-Office/FinancialAlignmentInitiative/Downloads/MMPPerformanceDataTechNotesCY2017_06262017.pdf [Last accessed 18 November 2017]
- Dunlay SM, Eveleth JM, Shah ND, et al. Medication adherence among community-dwelling patients with heart failure. Mayo Clin Proc 2011;86:273-81
- Huber CA, Rapold R, Brüngger B, et al. One-year adherence to oral antihyperglycemic medication and risk prediction of patient outcomes for adults with diabetes mellitus: an observational study. Medicine (Baltimore) 2016;95:e3994
- Hedna K, Hakkarainen KM, Gyllensten H, et al. Adherence to antihypertensive therapy and elevated blood pressure: should we consider the use of multiple medications? PLoS One 2015;10(9):e0137451
- Bureau of Labor Statistics. Databases, Tables & Calculators by Subject: Inflation and Prices. Washington (DC): BLS, 2017. Available at: http://www.bls.gov/data/#prices [Last accessed 18 November 2017]
- Chandra A. Prevalence of hypothyroidism in patients with chronic kidney disease: a cross-sectional study from North India. Kidney Res Clin Pract 2016;35:165-8
- Furukawa S, Yamamoto S, Todo Y, et al. Association between subclinical hypothyroidism and diabetic nephropathy in patients with type 2 diabetes mellitus. Endocr J 2014;61:1011-18
- Lisotto C, Mainardi F, Maggioni F, et al. The comorbidity between migraine and hypothyroidism. J Headache Pain 2013;14:P138
- Garber JR, Cobin RH, Gharib H, et al. Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocr Pract 2012;18:988-1028
- Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005;43:1130-9
- Wong ES, Bryson CL, Hebert PL, et al. Estimating the impact of oral diabetes medication adherence on medical costs in VA. Ann Pharmacother 2014;48:978-85
- Simoni-Wastila L, Wei YJ, Qian J, et al. Association of chronic obstructive pulmonary disease maintenance medication adherence with all-cause hospitalization and spending in a Medicare population. Am J Geriatr Pharmacother 2012;10:201-10
- Khanna R, Pace PF, Mahabaleshwarkar R, et al. Medication adherence among recipients with chronic diseases enrolled in a state Medicaid program. Popul Health Manag 2012;15:253-60
- National Center for Health Statistics. Health, United States, 2016: with chartbook on long-term trends in health. Hyattsville, MD, 2017. Available at: https://www.cdc.gov/nchs/data/hus/hus16.pdf [Last accessed 9 April 2018]
- Bodenheimer T, Wagner EH, Grumbach K. Improving primary care for patients with chronic illness: the chronic care model, Part 2. JAMA 2002;288:1909-14
- American Diabetes Association. Economic costs of diabetes in the U.S. in 2012. Diab Care 2013;36:1033-46
- Sydnor ER, Perl TM. Hospital epidemiology and infection control in acute-care settings. Clin Microbiol Rev 2011;24:141-73
- Quach C, McArthur M, McGeer A, et al. Risk of infection following a visit to the emergency department: a cohort study. CMAJ 2012;184:E232-9
- Gleason KM, McDaniel MR, Feinglass J, et al. Results of the medications at transitions and clinical handoffs (MATCH) study: an analysis of medication reconciliation errors and risk factors at hospital admission. J Gen Intern Med 2010;25:441-7
- Bleetman A, Sanusi S, Dale T, et al. Human factors and error prevention in emergency medicine. Emerg Med J 2012;29:389-93
- Joynt KE, Gawande AA, Orav EJ, et al. Contribution of preventable acute care spending to total spending for high-cost Medicare patients. JAMA 2013;309:2572-8
- McPhail SM. Multimorbidity in chronic disease: impact on health care resources and costs. Risk Manag Healthc Policy 2016;9:143-56
- Corsonello A, Pedone C, Garasto S, et al. The impact of comorbidity on medication adherence and therapeutic goals. BMC Geriatr 2010;10:L21
- Mohammed S, El-Menyar A, Abdulkari S, et al. Impact of polypharmacy on adherence to evidence-based medication in patients who underwent percutaneous coronary intervention. Curr Vasc Pharmacol 2016;14:388-93
- Garber AJ, Moghissi ES, Bransome ED Jr, et al. American College of Endocrinology position statement on inpatient diabetes and metabolic control. Endocr Pract 2004;10:77-82
- Oomen HA, Schipperijn AJ, Drexhage HA. The prevalence of affective disorder and in particular of a rapid cycling of bipolar disorder in patients with abnormal thyroid function tests. Clin Endocrinol (Oxf) 1996;45:215-23
- Bathla M, Singh M, Relan P. Prevalence of anxiety and depressive symptoms among patients with hypothyroidism. Indian J Endocrinol Metab 2016;20:468-74
- Knudsen N, Laurberg P, Rasmussen LB, et al. Small differences in thyroid function may be important for body mass index and the occurrence of obesity in the population. J Clin Endocrinol Metab 2005;90:4019-24
- Kleiner J, Altshuler L, Hendrick V, et al. Lithium-induced subclinical hypothyroidism: review of the literature and guidelines for treatment. J Clin Psychiatry 1999;60:249-55
- Baumgartner A. Thyroxine and the treatment of affective disorders: an overview of the results of basic and clinical research. Int J Neuropsychopharmacol 2000;3:149-65
- Joffe RT. Hormone treatment of depression. Dialogues Clin Neurosci 2011;13:127-38
- Cappola AR, Fried LP, Arnold AM, et al. Thyroid status, cardiovascular risk, and mortality in older adults. JAMA 2006;295:1033-41
- Gammage MD, Parle JV, Holder RL, et al. Association between serum free thyroxine concentration and atrial fibrillation. Arch Intern Med 2007;167:928-34
- Rodondi N, Bauer DC, Cappola AR, et al. Subclinical thyroid dysfunction, cardiac function, and the risk of heart failure: the cardiovascular health study. J Am Coll Cardiol 2008;52:1152-9
- Biondi B. Mechanisms in endocrinology: heart failure and thyroid dysfunction. Eur J Endocrinol 2012;167:609-18
- Gluvic Z, Sudar E, Tica J, et al. Effects of levothyroxine replacement therapy on parameters of metabolic syndrome and atherosclerosis in hypothyroid patients: a prospective pilot study. Int J Endocrinol 2015:147070
- Lioudaki E, Mavroeidi NG, Mikhailidis DP, Ganatakis ES. Subclinical hypothyroidism and vascular risk: an update. Hormones 2013;12:495-506