Impact of UK Primary Care Policy Reforms on Short-Stay Unplanned Hospital Admissions for Children With Primary Care–Sensitive Conditions

Main Findings

Our findings confirm an association between the timing of major national primary care policy reform in the United Kingdom in 2004 and increases in short-stay admission rates for PCS chronic conditions in children. The 8% rise in 2004, above the increasing trend, was equivalent to 8,500 extra short-stay admissions. These increases were accompanied by falls in admissions of children referred by a primary care physician. We did not find an immediate change in the level of short-stay admission rates for infectious illness, which were increasing annually by 5% annually well before the reforms of 2004 against a steadily rising trend for injury admissions of 1% to 2% annually.

Strengths and Weaknesses in Relation to Previous Research

Our study is the first to use interrupted time series analysis to investigate the timing of primary care policy reforms on PCS admissions among children. Our study’s strengths include its size and national population coverage. National administrative admission data are a useful tool for exploring the impact of national policy change as they are not subject to bias due to regional variations of individual clinicians.²¹ Our definition of a PCS admission was deliberately broad and captured a wide spectrum of potentially avoidable admissions. Pediatric indicator sets adopted by previous studies^10,22 accounted for only selected infectious and chronic conditions that, together, explained less than 30% of short-stay admissions in children. By contrast, in our study, the 10 most frequent causes for admission for potentially avoidable admissions for PCS infectious and chronic conditions accounted for more than 70% of short-stay admissions (Supplemental Appendix 3, http://www.annfammed.org/content/13/3/214/suppl/DC1). Our broader composite indicators for acute and chronic PCS conditions may be useful for future research to monitor the impact of primary care on potentially avoidable admissions among children.

We acknowledge that our results are subject to a number of limitations, however. Inherent biases exist in the use of clinical codes and length of stay to derive a proxy for identifying PCS admissions. Changes in accuracy and completeness of clinical coding could have affected diagnosis categorization or referral information over the study period,³ although the use of broadly defined diagnosis groups is less likely to be affected by variations in coding of individual conditions.²³ As we present only 4 years of data before policy reform, it is possible that a more important underlying trend could have been obscured; however, our previous report of a 41% increase in short-stay admissions over a 10-year period (between 1997 and 2006) is comparable to our findings of a pre-2004 rate of 4% per year.^7,8

Our findings may be confounded by concurrent health policy and service changes in hospital care around 2004 (Table 1). UK government performance targets introduced in 2000 were tightened in 2004 whereby emergency departments incurred heavy penalties if patients waited longer than 4 hours. In response, many hospitals introduced short-stay admission facilities for assessment and observation, which may explain increases in same-day discharge rates but not differing trends across diagnosis categories.^2,10,24 Financial tariff schemes for hospital activities including admissions were introduced starting in 2003, but there was only limited evidence of “gaming” by a small number of hospital trusts to increase payments.^3,25 Shorter length of stay is associated with improved efficiency in hospitals when offset by a reduction in longer-stay admissions of 2 days or more; we adjusted for changes in long-stay admission rates and found increased efficiency does not explain the increases in short-stay admissions.

Implications and Future Research

Although we cannot infer causation from our findings, the magnitude of an 11% increase in short-stay admissions for chronic disease lends weight to speculation from previous reports that such admissions may increase when primary care provision is withdrawn.¹⁰ Our findings that short-stay admission rates among children with chronic conditions changed immediately in 2004 and are now surpassing other causes of admission in older children are particularly concerning. This development may indicate an adverse impact of financial incentive schemes focusing on chronic conditions in adults.²⁶ The success of the scheme on improving quality of care and clinical outcomes has been questioned,^27,28 and there has been concern that it has widened inequalities as a result of unintended consequences for children and other groups.^29,30 Continuity of care, important in building a parent-physician relationship and preventing the use of emergency health services, also reportedly suffered with the introduction of pay for performance.^31–33 Evidence of the impact of pay for performance in the United Kingdom on children’s access to primary care physicians is limited, but primary care consultation rates rose between 1995 and 2006 in all age-groups except for children aged 5 to 14 years.³⁴

We did not find a direct relationship for rising admissions for infectious illness in children, which predated the primary care policy reforms. Infectious illness is common among young children and particularly sensitive to the availability of out-of-hours care. Steep increases before 2004 may reflect gradual changes in out-of-hours primary care provision from 2000³⁵; by 2004, more than 95% of primary care physicians had opted out of this care provision nationally. Certainly, the large increases in admissions with same-day discharge may reflect a trend in emergency departments toward hospitalizing children for observation.^10,36

The expansion in primary care–based activity as a result of reforms has meant that primary care physicians are struggling to maintain their existing workload,³⁷ and existing health care provision in England is not sufficient to contain childhood illness in the community. Concerns about rising emergency department visits and short-stay admissions have sparked calls for a reversal of policy reform, suggesting UK primary care physicians should extend access to primary care outside of core hours, and controversial pilot schemes for 24/7 care are now proposed in many areas.³⁸ Such changes, however, have huge implications for the primary care physician workforce and UK health system budgets. We recommend further research to model clinical and cost-effectiveness and survey parental attitudes to weigh best options for avoiding admissions in children.