Cynthia A. Beck

Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data.

Objectives:Implementation of the International Statistical Classification of Disease and Related Health Problems, 10th Revision (ICD-10) coding system presents challenges for using administrative data. Recognizing this, we conducted a multistep process to develop ICD-10 coding algorithms to define Charlson and Elixhauser comorbidities in administrative data and assess the performance of the resulting algorithms. Methods:ICD-10 coding algorithms were developed by “translation” of the ICD-9-CM codes constituting Deyos (for Charlson comorbidities) and Elixhausers coding algorithms and by physicians’ assessment of the face-validity of selected ICD-10 codes. The process of carefully developing ICD-10 algorithms also produced modified and enhanced ICD-9-CM coding algorithms for the Charlson and Elixhauser comorbidities. We then used data on in-patients aged 18 years and older in ICD-9-CM and ICD-10 administrative hospital discharge data from a Canadian health region to assess the comorbidity frequencies and mortality prediction achieved by the original ICD-9-CM algorithms, the enhanced ICD-9-CM algorithms, and the new ICD-10 coding algorithms. Results:Among 56,585 patients in the ICD-9-CM data and 58,805 patients in the ICD-10 data, frequencies of the 17 Charlson comorbidities and the 30 Elixhauser comorbidities remained generally similar across algorithms. The new ICD-10 and enhanced ICD-9-CM coding algorithms either matched or outperformed the original Deyo and Elixhauser ICD-9-CM coding algorithms in predicting in-hospital mortality. The C-statistic was 0.842 for Deyos ICD-9-CM coding algorithm, 0.860 for the ICD-10 coding algorithm, and 0.859 for the enhanced ICD-9-CM coding algorithm, 0.868 for the original Elixhauser ICD-9-CM coding algorithm, 0.870 for the ICD-10 coding algorithm and 0.878 for the enhanced ICD-9-CM coding algorithm. Conclusions:These newly developed ICD-10 and ICD-9-CM comorbidity coding algorithms produce similar estimates of comorbidity prevalence in administrative data, and may outperform existing ICD-9-CM coding algorithms.

Major depression in multiple sclerosis: a population-based perspective.

Objective: To determine the prevalence of major depression in multiple sclerosis (MS) in a population-based sample controlling for nonspecific illness effects. Methods: This study used data from a large-scale national survey conducted in Canada: the Canadian Community Health Survey (CCHS). The analysis included 115,071 CCHS subjects who were 18 years or older at the time of data collection. The CCHS interview obtained self-reported diagnoses of MS and employed a brief predictive interview for major depression: the Composite International Diagnostic Interview Short Form for Major Depression. The 12-month period prevalence of major depression was estimated in subjects with and without MS and with and without other long-term medical conditions. Results: The prevalence of major depression was elevated in persons with MS relative to those without MS and those reporting other conditions. The association persisted after adjustment for age and sex (adjusted odds ratio = 2.3, 95% CI 1.6 to 3.3). Major depression prevalence in MS for those in the 18- to 45-year age range was high at 25.7% (95% CI 15.6 to 35.7). Conclusions: The prevalence of major depression in the population with MS is elevated. This elevation is not an artifact of selection bias and exceeds that associated with having one or more other long-term conditions.

Descriptive Epidemiology of Major Depression in Canada

Objective: The Canadian Community Health Survey: Mental Health and Well-Being (CCHS 1.2) is the first national study to use a full version of the Composite International Diagnostic Interview. For this reason, and because of its large sample size, the CCHS 1.2 is capable of providing the best currently available description of major depression epidemiology in Canada. Using the CCHS 1.2 data, our study aimed to describe the epidemiology of major depression in Canada. Method: All estimates used appropriate sampling weights and bootstrap variance estimation procedures. The analysis consisted of estimating proportions supplemented by logistic regression modelling. Results: The lifetime prevalence of major depressive episode was 12.2%. Past-year episodes were reported by 4.8% of the sample; 1.8% reported an episode in the past 30 days. As expected, major depression was more common in women than in men, but the difference became smaller with advancing age. The peak annual prevalence occurred in the group aged 15 to 25 years. The prevalence of major depression was not related to level of education but was related to having a chronic medical condition, to unemployment, and to income. Married people had the lowest prevalence, but the effect of marital status changed with age. Logistic regression analysis suggested that the annual prevalence may increase with age in men who never married. Conclusions: The prevalence of major depression in the CCHS 1.2 was slightly lower than that reported in the US and comparable to pan-European estimates. The pattern of association with demographic and clinical variables, however, is broadly similar. An increasing prevalence with age in single (never-married) men was an unexpected finding.

Long-Term Medical Conditions and Major Depression: Strength of Association for Specific Conditions in the General Population

Background: The prevalence of major depression (MD) in persons with nonpsychiatric medical conditions is an indicator of clinical need in those groups, an indicator of the feasibility of screening and case-finding efforts, and a source of etiologic hypotheses. This analysis explores the prevalence of MD in the general population in relation to various long-term medical conditions. Methods: We used a dataset from a large-scale Canadian national health survey, the Canadian Community Health Survey (CCHS). The sample consisted of 115 071 subjects aged 18 years and over, randomly sampled from the Canadian population. The survey interview recorded self-reported diagnoses of various long-term medical conditions and employed a brief predictive interview for MD, the Composite International Diagnostic Interview Short Form for Major Depression. Logistic regression was used to adjust estimates of association for age and sex. Results: The conditions most strongly associated with MD were chronic fatigue syndrome (adjusted odds ratio [AOR] 7.2) and fibromyalgia (AOR 3.4). The conditions least strongly associated were hypertension (AOR 1.2), diabetes, heart disease, and thyroid disease (AOR 1.4 in each case). We found associations with various gastrointestinal, neurologic, and respiratory conditions. Conclusions: A diverse set of long-term medical conditions are associated with MD, although previous studies might have lacked power to detect some of these associations. The strength of association in prevalence data, however, varies across specific conditions.

Exploring physician specialist response rates to web-based surveys

BackgroundSurvey research in healthcare is an important tool to collect information about healthcare delivery, service use and overall issues relating to quality of care. Unfortunately, physicians are often a group with low survey response rates and little research has looked at response rates among physician specialists. For these reasons, the purpose of this project was to explore survey response rates among physician specialists in a large metropolitan Canadian city.MethodsAs part of a larger project to look at physician payment plans, an online survey about medical billing practices was distributed to 904 physicians from various medical specialties. The primary method for physicians to complete the survey was via the Internet using a well-known and established survey company (www.surveymonkey.com). Multiple methods were used to encourage survey response such as individual personalized email invitations, multiple reminders, and a draw for three gift certificate prizes were used to increase response rate. Descriptive statistics were used to assess response rates and reasons for non-response.ResultsOverall survey response rate was 35.0%. Response rates varied by specialty: Neurology/neurosurgery (46.6%); internal medicine (42.9%); general surgery (29.6%); pediatrics (29.2%); and psychiatry (27.1%). Non-respondents listed lack of time/survey burden as the main reason for not responding to our survey.ConclusionsOur survey results provide a look into the challenges of collecting healthcare research where response rates to surveys are often low. The findings presented here should help researchers in planning future survey based studies. Findings from this study and others suggest smaller monetary incentives for each individual may be a more appropriate way to increase response rates.

Regional variation of multiple sclerosis prevalence in Canada

Objective: To describe the regional distribution of multiple sclerosis (MS) prevalence in Canada, controlling for age and sex. Methods: This study used data from the Canadian Community Health Survey, a large general health survey (n=131,535) conducted in 2000/2001. Subjects aged 18 and over were included in the current analysis (n=116,109). The presence of MS was determined by self-report. Prevalence was computed in five regions (Atlantic, Quebec, Ontario, Prairies and British Columbia). Logistic regression was used to compare regions and examine for confounding/interaction by age and sex. Results: The overall Canadian MS prevalence was 240 per 100 000 (95%CI: 210-280). Prevalence ranged from 180 (95%CI: 90-260) in Quebec to 350 (95%CI: 230-470) in Atlantic Canada. Logistic regression revealed no statistical difference between the odds of MS in Quebec, Ontario and British Columbia adjusted for age and sex. The adjusted odds of MS in the Prairies and Atlantic regions were significantly higher than in the other regions combined, with odds ratios of 1.7 (95%CI: 1.1-2.4, p<0.01) and 1.6 (95%CI: 1.1-2.4, p<0.05) respectively. Sensitivity analysis demonstrated similar prevalence in the nonaboriginal/nonimmigrant group (n=96 219). Conclusion: Results suggest that Canadian MS prevalence differs by region. If validated, these regional differences may facilitate investigation of environmental influences.

Recommendations for optimal ICD codes to study neurologic conditions A systematic review

Objective: Administrative health data are frequently used for large population-based studies. However, the validity of these data for identifying neurologic conditions is uncertain. Methods: This article systematically reviews the literature to assess the validity of administrative data for identifying patients with neurologic conditions. Two reviewers independently assessed for eligibility all abstracts and full-text articles identified through a systematic search of Medline and Embase. Study data were abstracted on a standardized abstraction form to identify ICD code–based case definitions and corresponding sensitivity, specificity, positive predictive values (PPVs), and negative predictive values (NPVs). Results: Thirty full-text articles met the eligibility criteria. These included 8 studies for Alzheimer disease/dementia (sensitivity: 8–86.5, specificity: 56.3–100, PPV: 60–97.9, NPV: 68.0–98.9), 2 for brain tumor (sensitivity: 54.0–100, specificity: 97.0–99.0, PPV: 91.0–98.0), 4 for epilepsy (sensitivity: 98.8, specificity: 69.6, PPV: 62.0–100, NPV: 89.5–99.1), 4 for motor neuron disease (sensitivity: 78.9–93.0, specificity: 99.0–99.9, PPV: 38.0–90.0, NPV: 99), 2 for multiple sclerosis (sensitivity: 85–92.4, specificity: 55.9–92.6, PPV: 74.5–92.7, NPV: 70.8–91.9), 4 for Parkinson disease/parkinsonism (sensitivity: 18.7–100, specificity: 0–99.9, PPV: 38.6–81.0, NPV: 46.0), 3 for spinal cord injury (sensitivity: 0.9–90.6, specificity: 31.9–100, PPV: 27.3–100), and 3 for traumatic brain injury (sensitivity: 45.9–78.0 specificity: 97.8, PPV: 23.7–98.0, NPV: 99.2). No studies met eligibility criteria for cerebral palsy, dystonia, Huntington disease, hydrocephalus, muscular dystrophy, spina bifida, or Tourette syndrome. Conclusions: To ensure the accurate interpretation of population-based studies with use of administrative health data, the accuracy of case definitions for neurologic conditions needs to be taken into consideration.

Antidepressant utilization in Canada

ObjectiveAntidepressant utilization can be used as an indicator of appropriate treatment for major depression. The objective of this study was to characterize antidepressant utilization in Canada, including the relationships of antidepressant use with sociodemographic variables, past-year and lifetime depression, number of past depressive episodes, and other possible indications for antidepressants.MethodWe examined data from the Canadian Community Health Survey (CCHS) Cycle 1.2. The CCHS was a nationally representative mental health survey (N=36,984) conducted in 2002 that included a diagnostic instrument for past-year and lifetime major depressive episodes and other psychiatric disorders and a record of past-year antidepressant use.ResultsOverall, 5.8% of Canadians were taking antidepressants, higher than the annual prevalence of major depressive episode (4.8%) in the survey. Among persons with a past-year major depressive episode, the frequency of antidepressant use was 40.4%. After application of adjustments for probable successful outcomes of treatment, the estimated frequency of antidepressant use for major depression was more than 50%. Frequency of antidepressant treatment among those with a history of depression but without a past-year episode increased with the number of previous episodes. Among those taking antidepressants over the past year, only 33.1% had had a past-year episode of major depression. Migraine, fibromyalgia, anxiety disorder, or past depression was present in more than 60% of those taking antidepressants without a past-year episode of depression.ConclusionsThe CCHS results suggest that antidepressant use has increased substantially since the early 1990s, and also that these medications are employed extensively for indications other than depression.

Comorbidity of major depression with substance use disorders.

Objectives: In the Canadian adult population, we aimed to 1) estimate the 12-month prevalence of major depressive disorder (MDD) in persons with a diagnosis of harmful alcohol use, alcohol dependence, and drug dependence; 2) estimate the 12-month prevalence of harmful alcohol use, alcohol dependence, and drug dependence in persons with a 12-month and lifetime diagnosis of MDD; 3) identify socioeconomic correlates of substance use disorder–major depression comorbidity; 4) determine how comorbidity impacts the prevalence of suicidal thoughts; and 5) determine how comorbidity affects mental health care used. Methods: We examined data from the Canadian Community Health Survey: Mental Health and Well-Being (CCHS 1.2). Results: The 12-month prevalences of MDD in persons with a substance use disorder (SUD) were 6.9% for harmful alcohol use (95% confidence interval [CI], 5.2 to 8.5), 8.8% for alcohol dependence (95%CI, 6.6 to 11.0), and 16.1% for drug dependence (95%CI, 10.3 to 21.9). Conversely, the 12-month prevalences of harmful alcohol use, alcohol dependence, and drug dependence in persons with a 12-month diagnosis of MDD were 12.3% (95%CI, 9.4 to 15.2), 5.8% (95%CI, 4.3 to 7.3), and 3.2% (95%CI, 2.0 to 4.4), respectively. Regression modelling did not identify any socioeconomic predictors of SUD–MDD comorbidity. Substance dependence and MDD independently predicted higher prevalence of suicidal thoughts and mental health treatment use. Conclusions: SUDs cooccur with a high frequency in cases of MDD. Clinicians and mental health services should consider routine assessment of SUDs in depression patients.

Psychotropic Medication Use in Canada

Background: Psychotropic medication use can be employed as an indicator of appropriate treatment for mental disorders. The Canadian Community Health Survey: Mental Health and Well-Being (CCHS 1.2) offers the first opportunity to characterize Canadian psychotropic medication use on a national level within diagnostic groups as assessed by a full version of the Composite International Diagnostic Interview (CIDI). Method: We assessed the prevalence of antidepressant, sedative-hypnotic, mood stabilizer, psychostimulant, and antipsychotic use over 2 days overall and in subgroups defined by CIDI-diagnosed disorders and demographics. We employed sampling weights and bootstrap methods. Results: Overall psychotropic drug utilization was 7.2%. Utilization was higher for women and with increasing age. With any lifetime CIDI-diagnosed disorder assessed in the CCHS 1.2, utilization was 19.3%, whereas without such disorders, it was 4.1%. Selective serotonin reuptake inhibitors (SSRIs) were the most commonly used antidepressants for those with a past-year major depressive episode (17.8%), followed by venlafaxine (7.4%). Among people aged 15 to 19 years, antidepressant use was 1.8% overall and 11.7% among those with past-year depression; SSRIs made up the majority of use. Sedative-hypnotics were used by 3.1% overall, increasing with age to 11.1% over 75 years. Conclusions: International comparison is difficult because of different evaluation methods, but antidepressant use may be higher and antipsychotic use lower in Canada than in recent European and American reports. In light of the relative lack of contemporary evidence for antidepressant efficacy in adolescents, it is likely that antidepressant use among those aged 15 to 19 years will continue to decline. The increased use of sedative-hypnotics with age is of concern, given the associated risk of adverse effects among seniors.