What is the impact of a pay-for-performance scheme to give long acting reversible contraception (LARC) advice in general practice on individual women in terms of contraceptive prescriptions and abortions?

Problem

The UK government introduced a pay-for-performance (P4P) scheme which linked GP income to performance targets in 2004. Targets were introduced in 2009/10 that remunerated GPs to give LARC advice to women aged 13 to 54 years, with the aim of offering the best choice of contraception to reduce unplanned and unwanted pregnancies.

Our published study using interrupted time series analysis reported the P4P scheme was associated with 13% more LARC prescriptions and 38% fewer abortions than expected. This impact was greater in younger women and those from more disadvantaged backgrounds.

Ecological studies cannot infer association at the individual level. Our aim is to examine the impact of LARC advice given to individual women on LARC prescription and abortion.

Approach

We constructed a cohort of women using Clinical Practice Research Datalink (CPRD) who were aged at least 13 and no older than 54 years from financial year 2009/10, the year in which the P4P was introduced. This includes a population of over 3 million women from over 600 practices across the UK.

There were two eligible groups of women for LARC advice: those prescribed non-LARC hormonal contraception (NLHC), and emergency hormonal contraception (EHC). Comparison groups included women who were eligible for but did not receive LARC advice, women not eligible for LARC advice, and women who were ineligible but given LARC advice.

We used documented LARC advice as the intervention. Our outcome measures were LARC prescriptions and abortion events within 18 months after a woman became eligible for LARC advice. Our study period spanned five financial years from 2009/10 to 2013/14.

We stratified the groups by age and deprivation. We assigned women to one of 4 age groups: 13 to 19, 20 to 24, 25 to 34, and 35 to 54 years. Women were assigned to one of 5 deprivation groups based on their residential postcode using English Indices of Multiple Deprivation (IMD) quintiles that range from IMD 1 (least deprived) to IMD 5 (most deprived); IMD data are available for English populations only

Findings

Using multinomial logistic regression, we will report the odds of categorial events (prescription of LARC and abortion) among the four groups of women stratified by age group and IMD.

We will use model fit statistics to check the outputs. To look for outliers or influential data points, we will run separate logit models and use the diagnostics tools on each model.

Consequences

If the results confirm the findings from our earlier study, this will strengthen evidence base that P4P scheme to give LARC advice increased LARC prescriptions and reduced abortions, and had greater impact in younger women and those from more disadvantaged backgrounds, and suggests potential to reduce health inequalities by improved awareness of options for effective fertility control.