Table 2.
Clusters within the high-risk population
| Cluster |
|||||
|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | ANOVA/Kruskal- Wallis/χ2 test | |
| Clustering variables | |||||
| Number of emergency hospitalisations over 2008–2011, median (IQR) | 1 (0–1)‡ | 3 (2–4)* | 1 (0–1)‡ | 3 (2–4)* | KW: <0.000 |
| Number of non-emergency hospitalisations over 2008–2011, median (IQR) | 3 (2–5)* | 3 (2–5)* | 0 (0–1)* | 0 (0–1)* | KW: <0.000 |
| Number of outpatient attendances over 2008–2011, median (IQR) | 24 (16–38)* | 29 (18–46)* | 7 (3–13)* | 10 (5–18)* | KW: <0.000 |
| Number of GP visits over 2008–2011, median (IQR) | 61 (43–90)* | 57 (40–86)* | 55 (35–82)* | 42 (26–65)* | KW: <0.000 |
| Post hoc analysis of other variables | |||||
| Number of people | 1967 | 1807 | 1831 | 1861 | |
| Predicted proportion with any emergency hospitalisations in 2012 (based on average risk score), % | 21* | 38* | 20* | 31* | AN: <0.000 |
| Actual proportion with any emergency hospitalisations in 2012, % | 19‡ | 35‡ | 21‡ | 34‡ | χ2: <0.000 |
| Age at end of study period, mean | 79* | 67* | 83* | 71* | AN: <0.000 |
| Number of long-term conditions, median (IQR) | 2 (1–3)‡ | 2 (1–3)‡ | 1 (1–2)* | 1 (1–2)* | KW: <0.000 |
| Number of emergency hospitalisations in 2012, median (IQR) | 2 (1–3)‡ | 2 (1–3)‡ | 1 (1–2)‡ | 1 (1–2)‡ | KW: <0.000 |
| Number of non-emergency hospitalisations in 2012, median (IQR) | 0 (0–0)‡ | 0 (0–1)‡ | 0 (0–0)* | 0 (0–1)* | KW: <0.000 |
| Number of outpatient attendances in 2012, median (IQR) | 0 (0–1)* | 0 (0–1)* | 0 (0–0)* | 0 (0–0)* | KW: <0.000 |
| Number of GP visits in 2012, median (IQR) | 5 (2–10)* | 6 (3–11)‡ | 2 (0–4)‡ | 2 (0–5)* | KW: <0.000 |
| Prevalence of acute myocardial infarction, % | 15* | 23* | 10* | 19* | χ2: <0.000 |
| Prevalence of asthma, % | 28† | 26 | 24† | 25 | χ2: 0.028 |
| Prevalence of cancer, % | 26* | 22* | 8* | 5* | χ2: <0.000 |
| Prevalence of cerebrovascular disease, % | 9‡ | 15‡ | 10‡ | 18‡ | χ2: <0.000 |
| Prevalence of congestive heart failure, % | 8* | 13‡ | 5* | 13‡ | χ2: <0.000 |
| Prevalence of COPD, % | 18† | 17† | 13* | 18† | χ2: <0.000 |
| Prevalence of dementia, % | 3‡ | 3‡ | 5‡ | 7‡ | χ2: <0.000 |
| Prevalence of diabetes, % | 28‡ | 22‡ | 28‡ | 22‡ | χ2: <0.000 |
| Prevalence of HIV/AIDS, % | 0 | 0 | 0 | 0 | χ2: 0.39 |
| Prevalence of learning disabilities, % | 0† | 0† | 0 | 0 | χ2: 0.032 |
| Prevalence of liver disease, % | 1 | 1† | 0‡ | 1† | χ2: <0.000 |
| Prevalence of mental health conditions, % | 2† | 3† | 2† | 5* | χ2: <0.000 |
| Prevalence of paraplegia, % | 1‡ | 3‡ | 1‡ | 3‡ | χ2: <0.000 |
| Prevalence of peptic ulcer, % | 4† | 4† | 2‡ | 3 | χ2: <0.000 |
| Prevalence of peripheral vascular disease, % | 8* | 11* | 4‡ | 6‡ | χ2: <0.000 |
| Prevalence of renal disease, % | 23† | 23† | 24† | 18* | χ2: <0.000 |
| Prevalence of rheumatic disease, % | 10‡ | 8† | 6† | 5‡ | χ2: <0.000 |
*Significantly different from three other clusters.
†Significantly different from one other clusters.
‡Significantly different from two other clusters; all at 0.05/4=0.0125 significance level (Bonferroni adjustment).