Table 3

Actionable factors influencing report of unmet need (households with someone ill who did not seek medical attention) in 1999 and 2003 (n51633)


Crude OR
Adjusted OR
95%CI adjusted OR
Adjusted gain/1000
95%CI gain

Illiterate head of household
1.43
1.35
1.25–1.46
12.7
9.56–15.9
Female headed household
1.87
1.50
1.34–1.69
2.4
1.7–3.1
Rural resident
1.56
1.39
1.22–1.59
9.2
5.5–13.0
Poorer household (lowest 25 percentile 2003)
0.96
1.14
1.04–1.24
2.0
0.7–3.4
Unexplained difference between 2003 and 1999
3.41
3.48
3.2–3.77
36.8
34.5–39.2

1. The individual benefit is the adjusted Odds Ratio from logistic regression.

2. The PRI (proportion requiring intervention) is the proportion of service users who currently do not have the favourable value of the variable. For example, the proportion that currently does not get all the prescribed medicines is 80%.

3. The gain per 1000 is calculated by multiplying the PRI by the risk difference. This is the proportion who could potentially become satisfied with the service as a result of each intervention.

Cockcroft et al. Health Research Policy and Systems 2007 5:1   doi:10.1186/1478-4505-5-1