Table 2.
Summary of studies relating to cost and healthcare utilization for patients with MCC. Adapted from Lehnert (2011) (Lehnert et al., 2011).
| Study & country | Description & year | Impact |
|---|---|---|
| Healthcare costs | ||
| Fishman et al. (1997) United States (Fishman et al., 1997) |
Cross-sectional study with diagnostic and procedural data (1992) from Group Health Cooperative (GHC) of Pudget Sound (Western Washington State, U.S.) |
Each additional CC resulted in an expected increase in annual Healthcare costs (HCCs) of between 80% - 300%, depending on age, sex, and CC profile. |
| Hoffman et al. (1996) United States (Hoffman et al., 1996) |
Cross-sectional study with data from the 1987 National Medical Expenditure Survey (household component) | In comparison with elders with acute conditions only ($2713), those with one CC had annual HCCs about 1.8 times ($4887), and those with two or more CCs had costs about 3.6 times as high ($9881). |
| Crystal et al. (2000) United States (Crystal et al., 2000) |
Cross-sectional study with 1995 Medicare Current Beneficiary Survey data (use and cost files) |
The number of CCs was significantly and positively associated with total HCCs, annual OPE, and OPE as percentage of income (persons without CCs spent 13.8% of their income, those with five or more CCs 25.5%). |
| Hwang et al. (2001) United States (Hwang et al., 2001) |
Cross-sectional study with 1996 Medicare Expenditure Panel Survey data (household Component |
OPE increased with each additional CC and was about twice as high for elders with two CCs compared with those without CCs. This association was found for OPE for prescription drugs, home health, office visits, hospital use, and medical equipment but not for OPE for dental services and vision aids. |
| Physician usage | ||
| Hessel et al. (2000) Germany (Hessel et al., 2000) |
Cross-sectional study with data from a household survey by the University of Leipzig, Germany, March/April 1996 | The number of medical conditions was significantly and positively associated with the annual number of physician visits and number of medications taken on a daily basis (CCs were strongest predictor in each of the multiple regression analyses). |
| Bed utilization | ||
| Chan et al. (2002) Australia (Chan et al., 2002) |
Cross-sectional study with data from a household survey in the Randwick Municipality of Sydney (Australia), March 1998 to June 1999 | Multiple (three or more) CCs were a strong and significant predictor of emergency department admissions. |
| Ionescu-Ittu et al. (2007) Canada (Ionescu-Ittu et al., 2007) |
Cross-sectional study with random sample drawn from provincial administrative databases in Quebec, Canada, for 2000–2001 | Comorbidity was a significant independent predictor of emergency department use. In a multivariate analysis, comorbidity had a comparatively weak effect on emergency department use: One additional score on CCI increased the rate of emergency department use by 7%, one score on the CDS by 4%. |
| Landi et al. (2004) Italy (Landi et al., 2004) |
Observational cohort study with administrative data from six Italian home health care agencies (longitudinal data, 1997–2002) | Elders with any HA (at baseline) had significantly more CCs (3.9) than those without HA (3.2). In a multivariate analysis, elderly persons with five or more CCs were more than twice as likely to incur an HA, compared with those without CCs (during 1-year follow-up). |
| Librero et al. (1999) Spain (Librero et al., 1999) |
Cross-sectional study with administrative (hospital discharge) data from Valencia Health Service, Spain, 1993–1994 | Results from logistic regression with age comorbidity interaction: Patients aged 65 to 79 in the highest morbidity group (5+) had significantly lower chances of being hospitalized (OR 0.51) than those without CCs, whereas patients with moderate morbidity burden (1 to 2) had significantly higher chances (OR 1.24). |
| Condelius et al. (2008) Sweden (Condelius et al., 2008) |
Cross-sectional study with administrative registry data (2001) from four municipalities | In multivariate analyses, the number of CCs was significantly associated with acute and total number of admissions, and (less strongly) with planned HAs. |
| Condelius et al. (2008) Sweden (Condelius et al., 2008) |
Cross-sectional study with administrative registry data (2001) from four municipalities in southern Sweden | Elders with three or more HAs had significantly more CCs (3.45) than those with one (1.64) or two stays (2.61). |
| Chu and Pei (1999) Hong Kong (Chu & Pei, 1999) |
Prospective case–control study with emergency admissions (using administrative data) at Queen Mary Hospital of Hong Kong, 1996 | Compared with controls, readmission cases had significantly more CCs (3.1 vs.2.6). Number of CCs was a significant risk factor for early unplanned readmission in a multivariate analysis (OR 1.30). |
| Medication | ||
| Fahlman et al. (2006), United States (Fahlman et al., 2006) |
Retrospective review (crosssectional) of retail and mail order prescription claims data from Medicare + Choice (collected between January 1998 and December 2000), United States | Beneficiaries with higher numbers of comorbidities had significantly greater numbers of prescriptions (8 prescriptions for each additional comorbidity) and higher annual prescription drug expenditures and higher OPE. |