Kristina Johnell

A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena

Study objective: In social epidemiology, it is easy to compute and interpret measures of variation in multilevel linear regression, but technical difficulties exist in the case of logistic regression. The aim of this study was to present measures of variation appropriate for the logistic case in a didactic rather than a mathematical way. Design and participants: Data were used from the health survey conducted in 2000 in the county of Scania, Sweden, that comprised 10 723 persons aged 18–80 years living in 60 areas. Conducting multilevel logistic regression different techniques were applied to investigate whether the individual propensity to consult private physicians was statistically dependent on the area of residence (that is, intraclass correlation (ICC), median odds ratio (MOR)), the 80% interval odds ratio (IOR-80), and the sorting out index). Results: The MOR provided more interpretable information than the ICC on the relevance of the residential area for understanding the individual propensity of consulting private physicians. The MOR showed that the unexplained heterogeneity between areas was of greater relevance than the individual variables considered in the analysis (age, sex, and education) for understanding the individual propensity of visiting private physicians. Residing in a high education area increased the probability of visiting a private physician. However, the IOR showed that the unexplained variability between areas did not allow to clearly distinguishing low from high propensity areas with the area educational level. The sorting out index was equal to 82%. Conclusion: Measures of variation in logistic regression should be promoted in social epidemiological and public health research as efficient means of quantifying the importance of the context of residence for understanding disparities in health and health related behaviour.

The relationship between number of drugs and potential drug-drug interactions in the elderly: a study of over 600,000 elderly patients from the Swedish Prescribed Drug Register.

AbstractBackground: Drug-drug interactions (DDIs) are of great concern, as they are known to be related to adverse drug reactions and hospitalisations. In addition, many DDIs are regarded as predictable and avoidable; therefore, they may be considered as targets for education and interventions. Objective: To analyse the relationship between number of dispensed drugs and the probability of potential DDIs among the elderly by using the new Swedish Prescribed Drug Register. Methods: We analysed data on age, sex and dispensed drugs for people aged ≥75 years who were registered in the Swedish Prescribed Drug Register from October to December 2005, and constructed a list of current prescriptions for every individual on the arbitrarily chosen date of 31 December 2005. Thereafter, we included those who had at least two dispensed drugs to capture the elderly population at risk of being exposed to DDIs (n = 630 743). The main outcome measures were potentially clinically relevant DDIs (type C), which may require dose adjustment, and potentially serious DDIs (type D), which should be avoided. Results: The prevalence of type C potential DDIs was 26% and of type D potential DDIs 5% in the study population. There was a strong association between number of dispensed drugs and the probability of type C potential DDIs and an even stronger association for type D potential DDIs, after adjustment for age and sex. In addition, the probability of type D potential DDIs decreased with increasing age, and women had a lower probability of type D potential DDIs than men. Conclusion: There seems to be a strong relationship between number of dispensed drugs and potential DDIs, especially for potentially serious DDIs, which has implications for the importance of trying to minimise the number of drugs prescribed in the elderly. Our findings that the probability of potentially serious DDIs decreases with increasing age among the elderly and that elderly women have a lower probability of potentially serious DDIs than elderly men need to be verified and investigated by further research.

The Influence of Educational Level on Polypharmacy and Inappropriate Drug Use: A Register-Based Study of More Than 600,000 Older People

OBJECTIVES: To investigate whether low educational attainment is associated with polypharmacy and potential inappropriate drug use (IDU) in older people.

Trends in polypharmacy and potential drug-drug interactions across educational groups in elderly patients in Sweden for the period 1992 - 2002.

OBJECTIVE This study investigates the changes in drug use, polypharmacy and potential drug-drug interactions (DDIs) between educational groups of Swedish elderly over a 10-year period from 1992 - 2002. METHODS We used data from SWEOLD I (n = 512) from 1992 and SWEOLD II from 2002 (n = 561), which are nationally representative surveys of the elderly population in Sweden aged 77 years and older. Both community-based and institutionalized persons were included. Information on drug use was based on personal interviews and all drugs used in the two weeks prior to the studies were recorded. The three outcomes under study were drug use, polypharmacy (concurrent use of five or more drugs), and potential DDIs. RESULTS In the SWEOLD data from 1992 - 2002, the mean number of drugs used per person increased from 2.5 - 4.4. Overall, 81% of the study participants were drug users in 1992 as compared to 88% in 2002. The prevalence of polypharmacy increased 3-fold (from 18% in 1992 to 42% in 2002) after controlling for age and gender. In both SWEOLD surveys, the less educated reported polypharmacy more often (19% in 1992 and 46% in 2002) than the higher educated (12% in 1992 and 36% in 2002). Potential DDIs also increased, both among the less educated (14% in 1992 to 26% in 2002) and the higher educated (18% in 1992 to 24% in 2002). The most pronounced changes in the consumption of specific drug groups were observed in antithrombotic agents, beta-blocking agents, ACE inhibitors, and vitamin B12 and folic acid. In general, the use of most therapeutic classes increased more among the well educated compared to less educated men between 1992 and 2002, whereas the opposite relationship prevailed among women. CONCLUSION This study indicates that the use of drugs, polypharmacy and potential DDIs have increased during 1992 to 2002 among the elderly. These changes were most prominent among the less educated women. Polypharmacy and potential DDIs represent potential health hazards for the elderly. Therefore, the trends of increasing polypharmacy and drug-drug interactions deserve attention and the mechanisms behind should be investigated further.

Inappropriate Drug Use in the Elderly: A Nationwide Register-Based Study

Background: Potentially inappropriate drug use (IDU) is an important and preventable safety concern in the care of elderly patients and has been associated with adverse drug reactions, hospitalization, and mortality. Objective: To estimate the prevalence of potentially IDU among the elderly in Sweden and investigate whether age. sex, and number of dispensed drugs are associated with IDU. Methods: We analyzed data on age, sex, and dispensed drugs for people aged 75 years and older who were listed in the Swedish Prescribed Drug Register from October–December 2005 (N = 732 226). The main outcome measures of IDU were prescription of anticholinergics, prescription of long-acting benzodiazepines, concurrent use of 3 or more psychotropic drugs, and an indication of potentially serious drug-drug interactions. Results: Prevalence for IDU was 17%; for anticholinergic drugs 6%, long-acting benzodiazepines 5%, 3 or more psychotropic drugs 5%, and potentially serious drug–drug interactions 4%. After adjustment for age and sex, number of dispensed drugs was strongly associated with all 4 measures of IDU. After adjustment for sex and number of dispensed drugs, increasing age was moderately associated with a higher probability of IDU, long-acting benzodiazepines, and 3 or more psychotropic drugs, After adjustment for age and number of dispensed drugs, women had a slightly increased probability of IDU, anticholinergic drugs, long-acting benzodiazepines, and 3 or more psychotropic drugs. Conclusions: IDU was fairly common among the elderly in Sweden in 2005 and was strongly connected to the number of dispensed drugs they were taking. Older age and female sex were related to inappropriate use of psychotropic drugs, whereas the opposite relationship prevailed for potentially serious drug–drug interactions. Future research is needed to determine whether IDU will become more common due to increasing use of drugs among elderly persons. The challenge is to balance the problems related to IDU without denying older people potentially valuable drug therapy.

Pain Treatment in Elderly Persons With and Without Dementia : A population-Based Study of Institutionalized and Home-Dwelling Elderly

AbstractBackground: Several previous studies have reported an undertreatment of pain in elderly persons with dementia. It has also been suggested that persons with dementia may be at risk for inappropriate treatment of pain with psychotropics. Objectives: The objective of this study was to investigate if persons with dementia are as likely as persons without dementia to receive pharmacological pain treatment, after taking into account residential setting and pain-related disorders. We also aimed to investigate whether use of psychotropics is related to pain in persons with and without dementia. Methods: We used baseline data from the SNAC-K (Swedish National Study of Aging and Care — Kungsholmen). We analysed use of analgesics and psychotropics, prevalence of pain-related diagnoses, self-reported pain, dementia status and residential setting in 2610 participants aged >65 years. Results: Of the persons with dementia, 46% used at least one analgesic drug compared with 25% of those without dementia. Although persons with dementia reported pain less frequently than persons without dementia, the prevalence of pain-related diagnoses was similar. After adjustment for individual factors and residential setting (own home/institution), persons with dementia had a higher probability of use of paracetamol (acetaminophen) and psychotropics, whereas there were no significant differences in use of any analgesic, opioids and NSAIDs. Furthermore, having a pain-related diagnosis was associated with use of psychotropics in persons with dementia. Conclusions: Persons with dementia had a higher probability of use of paracetamol and were about as likely as persons without dementia to use any analgesic, opioids and NSAIDs, after adjustment for confounders. This may reflect a recent increased awareness of pain and pain management in persons with dementia, compared with previous studies that have reported an underuse of analgesics in persons with dementia. However, further research is needed to analyse if persons with dementia are appropriately treated for pain with regard to type of analgesic drug, pain intensity, indication, dosage and regimen.

Multi-dose drug dispensing and inappropriate drug use: A nationwide register-based study of over 700 000 elderly

Objective. To investigate whether the use of multi-dose drug dispensing is associated with potentially inappropriate drug use (IDU). Design. Cross-sectional nationwide register-based study. Setting. Sweden. Subjects. People aged ≥75 years registered in the Swedish Prescribed Drug Register during October–December 2005. Main outcome measures. Potentially IDU, i.e. anticholinergic drugs, long-acting benzodiazepines, concurrent use of three or more psychotropic drugs, and combinations of drugs that may lead to potentially serious drug–drug interactions (DDIs). Results. After adjustment for age and number of dispensed drugs, any IDU, anticholinergic drugs, and three or more psychotropic drugs were associated with multi-dose drug dispensing, whereas the opposite relationship prevailed for long-acting benzodiazepines among women and potentially serious DDIs in both men and women. Among the subjects aged 75–79 years, multi-dose drug dispensing was even more strongly associated with any IDU, anticholinergic drugs, three or more psychotropic drugs in both men and women, and long-acting benzodiazepines among men. Conclusion. Our results indicate that multi-dose users may be more exposed to potential IDU. However, multi-dose users seem to have a lower probability of potentially serious DDIs. Future research on multi-dose drug dispensing is necessary to ensure a high quality of drug therapy within this system.

Analysis of the association between polypharmacy and socioeconomic position among elderly aged ≥77 years in Sweden

BACKGROUND Polypharmacy is an important patient safety concern and has been associated with increased adverse drug reactions, hospitalization, and mortality in the elderly. However, few studies have analyzed the association between socioeconomic position (SEP) and the extent of drug use in older people. OBJECTIVES The aims of this study were to investigate the prevalence of polypharmacy and the association between polypharmacy and SEP, as measured by education, occupation, and income. METHODS For this cross-sectional study, data from a nationally representative sample (Swedish Panel Study of Living Conditions of the Oldest Old [SWEOLD] 2002) aged > or =77 years were analyzed (n = 621). Information on drug use was based on personal interviews, and polypharmacy was defined as concurrent use of > or =5 drugs. Other measurements included were age, sex, education, occupation, income, comorbidity, marital status, and living situation. The association between polypharmacy and SEP was assessed by logistic regression. RESULTS The mean number of drugs used in the sample was 4.4; in less educated patients, 4.6; more educated, 4.0. Polypharmacy was observed in 42.2% of the elderly. Overall, antithrombotic agents (42.5%), P-blocking agents (28.3%), and high-ceiling diuretics (28.0%) were the most prevalent drugs. Low education was associated with polypharmacy (odd ratio [OR], 1.46; 95% CI, 1.02-2.07), after controlling for age and sex. However, the association between low education level and polypharmacy was not significant after adjustment for age, sex, comorbidity, marital status, and living situation (OR, 1.39; 95% CI, 0.95-2.04). Moreover, we did not observe any association between occupation or income and polypharmacy. CONCLUSIONS The results from this study suggest that >40% of people aged > or =77 years in Sweden are exposed to polypharmacy, defined as the use of > or =5 drugs. There was a higher prevalence of polypharmacy among elderly with low education. However, after controlling for comorbidity, martial status, and living situation, polypharmacy was not related to low education. Further studies in larger populations are needed to elucidate the association between SEP and drug use.

The use of benzodiazpines and related drugs amongst older people in Sweden: Associated factors and concomitant use of other psychotropics

To determine the factors associated with use of benzodiazepines and benzodiazepine related drugs (BZDs/BZDRDs) in a nationwide population of older people.

Sex Differences in Inappropriate Drug Use: a Register-Based Study of Over 600,000 Older People

Background: Several studies have observed a higher rate of inappropriate drug use in elderly women than in elderly men, although few have focused explicitly on sex differences. Also, educational level (a measure of socioeconomic status) has been associated with both polypharmacy and inappropriate drug use in the elderly population, which could explain sex differences in inappropriate drug use. Objective: To investigate whether sex is associated with inappropriate drug use in a large general population of older people, after adjustment for age, socioeconomic status (ie, educational level), and comorbidity (ie, number of drugs). Methods: We conducted an analysis of data on sex, age, dispensed drugs, and education for people aged 75–89 years who were registered in the Swedish Prescribed Drug Register between July and October 2005 (N = 645,429). The main outcome measures were the filling of prescriptions for anticholinergic drugs or long-acting benzodiazepines, concurrent use of 3 or more psychotropic drugs, and potentially serious drug–drug interactions. Results: Potentially inappropriate drug use was more common in women (24.6%) than in men (19.3%). More women than men were exposed to all types of inappropriate drug use, except for potentially serious drug–drug interactions. Logistic regression analysis showed that women were more likely to be exposed to potentially inappropriate drug use than were men (OR 1.24; 95% CI 1.23 to 1.26) after adjustment for age, education, and number of drugs (used as a proxy for comorbidity). In particular, female sex was highly associated with Inappropriate psychotropic use (eg, long-acting benzodiazepines: adjusted OR [ORadjusted] 1.45; 95% CI 1.42 to 1.49; >3 psychotropic drugs: QRadjusted 1.50; 95% CI 1.47 to 1.53). In contrast, women had a lower probability of potentially serious drug-drug interactions than did men (ORadjusted 0.81; 95% CI 0.80 to 0.83). Conclusions: Inappropriate drug use in the elderly should be monitored separately for women and men. Our results suggest that neither socioeconomic status nor comorbidity seem to explain sex differences in inappropriate drug use. Future research in other large study populations is needed for investigation of other explanations behind these sex differences.