 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
11-February-2011, Vol 124 No 1329
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Continuity of care with general practitioners in
New Zealand: results from SoFIE-Primary Care
Santosh Jatrana, Peter Crampton, Ken Richardson
Continuity of care (COC) has been defined as seeing the same
health care provider over time, and is one of the four main domains of primary
care.1 Continuity of care presupposes the
existence of a regular source of care over time, regardless of the presence or
absence of disease or injury. It is intended to help the provider and the
patient build a long-term relationship in order to foster mutual trust between
provider and patient, and knowledge of both parties’ expectations and
needs.2
Studies, mainly from the US, have shown that increased
continuity of care is associated with positive health
outcomes,3 high quality
care,4 better medication and appointment
compliance, enhanced physician recognition of the patient’s health needs,
5and high patient satisfaction with the
care.6,7 Research has also suggested that
having a regular and consistent source of care is associated with lowering
health care costs 3 by decreasing use of
emergency services8,9 and
hospitalisations,10,11 particularly for
ambulatory-care-sensitive conditions (conditions that are more amenable to
primary care interventions).
The hypothesized benefits of continuity of care with a
primary care provider (PCP) are based on the accrued mutual knowledge, trust and
communication between patients and providers that arises from repeated
contact.1–3 Hence, there is considerable
policy interest in defining the characteristics of people who receive continuity
of care from their PCPs.
While the benefits of continuity of care with a primary care
provider are well documented, relatively little is known about those patients
who receive continuity of care. Studies which have looked at patients who did
not receive continuity of care noted that they were typically younger, female
and had relationship problems.12,13 Our goal in
this study is to construct a summary measure of continuity of care and to
enhance understanding of the prevalence of continuity of care in New Zealand
(NZ).
While defining the characteristics of those who receive
continuity of care is of interest in its own right, it is particularly important
in the NZ context, mainly because the different attributes of primary care have
not been studied to the same extent as in countries such as the US, leading to a
paucity of evidence that grounds the NZ experience in the international context.
Moreover, studies from the US mainly focus on a single State, clinic/provider or
hospital, or non-elderly population thus restricting the generality of the
results.
Other studies focus on continuity of care at the level of
the whole system, rather than at an individual patient level. This may, in part,
be due to the challenge of collecting information at the individual level about
aspects of primary care, or the inability of consumers to be valid judges of
primary care quality.14
This study extends the current literature on continuity of
care by using a large national survey and by including a variety of demographic,
socioeconomic, health behaviour and health variables. We hypothesise that those
who have greater need for care will experience a higher mean continuity of care
score.
MethodsDataThis research used SoFIE-Health data, which is part of
the Statistics New Zealand-led Survey of Family, Income and Employment (SoFIE).
SoFIE is a single fixed panel and is the largest longitudinal survey ever run in
New Zealand. It is a nationally representative study of about 22,000 adults,
drawn by random sampling of households, interviewed face-to-face. All adults in
the original sample are followed for a maximum duration of eight years starting
from October 2002, even if their household or family circumstances change. It
collects information once a year from the same individuals on income levels,
sources and changes; together with the major influences on income such as
employment and education experiences, household and family status and changes,
demographic factors and health status.
The SoFIE-Health module is comprised of 20 minutes of
questionnaire time in waves 3 (2004-05), 5 (2006-07) and 7 (2008-09), in the
following health-related domains: SF-36 (Short-Form health survey), Kessler-10
(K-10), perceived stress, chronic conditions (heart disease, diabetes, and
injury-related disability), tobacco smoking, alcohol consumption, health care
utilisation, access and continuity of primary health care, and an individual
deprivation score. The health module is administered to the original sample
members (OSM).
Main outcome variableThe main outcome measure used for this work was an
index of continuity of care which is assessed by the following four questions in
SoFIE-Health.
Q1: Would the same doctor or nurse take care
of you every time you go?
Q2: If you called them, could you talk to the person
that knows you best?
Q3: Do you think they know you very well as a
person?
Q4: Do you think they know what medical
problems are most important to you?
The response categories include definitely, probably,
probably not and definitely not and are coded/scored as 4,3,2,1 respectively so
that a higher total score indicates higher continuity of care. We based our
method on the Primary Care Assessment Tools (PCAT) in order to translate the
concept of continuity of care into characteristics that can be
measured.15,16
The Primary Care Assessment Tools were developed to
collect and analyse information needed to describe primary care services needed,
provided and experienced by the population. Following PCAT, we excluded those
individuals who refused to answer any of Q1 - Q4 above. Individuals who
responded “not sure, don’t remember” to more than 2 questions
were also excluded. For those who responded “not sure, don’t
remember” to only one of the 4 questions, we replaced “not sure,
don’t remember” with “probably”
The mean continuity of care score for an individual was
calculated by summing the score of the four questions for each individual and
dividing this sum by the number of questions (4 in this case). For a detailed
example of the creation of the score, see Jatrana et al
(2008a).17
Independent variablesWe included sociodemographic, health risk behaviour and
health status variables as covariates. Independent variables chosen for analyses
were based on our review of the literature and our research questions served as
a guide in the selection of variables to include in the model of continuity of
primary care. Sociodemographic variables in this analysis are age, gender,
marital status, ethnicity, family structure, household equivalised income,
labour force status, highest level of education achieved, NZDep (area
deprivation), and NZiDep (individual deprivation). Health behaviour and health
included current smoking status, Kesseler-10 and number of chronic conditions.
Categories for the various measures are shown in Table 1. A description of these
variables is as follows:
Age—Age
was calculated at the Wave 3 interview date and categorised into the following
age groups: 15-24, 25-44, 45-64, and 65+.
Ethnicity—This paper uses the
‘prioritised’ concept of ethnicity. With the
‘prioritised’ concept, each respondent was assigned to a mutually
exclusive ethnic group by means of a prioritisation system commonly used in New
Zealand: Māori, if any of the responses to self-identified ethnicity was
Māori; Pacific, if any one response was Pacific but not Māori; Asian,
if any one response was Asian but not Māori/Pacific; the remainder
non-Māori non-Pacific non-Asian (nMnPnA). The nMnPnA category mostly
comprises New Zealanders of European descent, but strictly speaking is not an
ethnic group.
Marital
status—Marital status relates to legal marital status and is
categorised into currently married, previously married
(separated/divorced/widowed) and never married.
NZDep2001—NZDep2001 is a
census-based small-area index of socioeconomic deprivation [24]. The Deprivation
index score of dwelling location is derived from NZDep and assigned to the small
area of the dwelling. NZDep2001 deprivation scores apply to areas
rather than individual people. The index scale used here is from 1 to 5, where 1
= the least deprived 20% of areas and 5 = the most deprived 20% of areas.
NZiDep—The
NZiDep index is a tool for measuring socioeconomic deprivation for individuals
and is based on eight simple questions which take about 2 minutes to administer
[25]. The final deprivation score was coded into the following five ordinal
categories. Relatively few people have the largest number of deprivation
characteristics.
1 = no deprivation characteristics
2 = one deprivation characteristic
3 = two deprivation characteristics
4 = three or four deprivation characteristics
5 = five or more deprivation characteristics
Income—In SoFIE, income is
collected from every individual over 15 years at every wave. Household income
was derived by totalling adult annual personal income (before tax) from all
sources received, consumer price index (CPI) adjusted for the quarter ending
December 2001 (the first reference quarter of the study), equivalised for
household economies of scale using a NZ-specific equivalisation index [26], and
categorized into tertiles: low (<$26,109), medium ($26,109 to $43,015) and
high (≥$43,016). For the analyses in this paper, equivalised household
income at wave 1 was used.
Education—The
education variable used in this analysis was the highest level of education at
Wave 3, categorised as no qualification, school qualification, and post-school
qualification.
Smoking—A
current smoking status variable was created from responses to questions
“Do you smoke cigarettes”, and “Have you ever been a regular
smoker” and is coded into three categories: current smoker, ex-smoker and
never smoker.
Kessler-10 scale—The Kessler-10
(K-10) is a scale measuring non-specific psychological distress [27, 28]. The
K-10 consists of ten questions about non-specific psychological distress and
seeks to measure the level of current anxiety and depressive symptoms based on
questions about negative emotional states a person may have experienced in the
four weeks prior to interview. The scores were grouped into four levels
according to the criteria developed by Andrews and Slade (2001): low (10-15),
moderate (16-21), high (22-29), and very high (30+) [29, 30].
Chronic diseases—As part of the
health module each respondent was asked “have you ever been told by a
doctor that you had”: Asthma, High Blood Pressure, High Cholesterol, Heart
Disease, Diabetes, Stroke, Migraines, Chronic Depression, Manic Depression or
Schizophrenia.
These data were coded into a co-morbidities index: 0,
1-2, >2 co-morbid diseases.
Statistical analysisThis paper provides cross-sectional analyses of wave 3
data. The sample used in the analyses consist of 18,320 adult (15 years and
above) OSMs. Analyses were carried out using means and 95% confidence intervals
(CI) to evaluate the bivariate associations between continuity of care scores
and other variables. Ordinary Least Square (OLS) regression was used to adjust
for covariates, including age, sex, marital status, ethnicity, household
equivalised income, labour force status, small area deprivation, individual
deprivation, education, smoking and health (self-assessed health, K-10 and
number of chronic conditions).
The population used in the regression analyses was
11,915 adult OSMs at wave 3 who had complete information on all the
socioeconomic, health behaviour and health characteristics. All counts presented
in this paper are random rounded (up or down) to the nearest multiple of 5, with
a minimum value of 10, as per the Statistics New Zealand confidentiality
protocol. All analyses were performed on unit level data using SAS version 8.2
within the Statistics New Zealand data laboratory.
ResultsTable 1 presents the associations between mean continuity of
care scores and demographics, socioeconomic, health and health behaviour
characteristics of the respondents. The mean score for continuity of care was
3.10 (95%CI: 3.09–3.11) with a range of 1.0–4.0. As the age of the
respondents increased, so did the mean continuity of care score, with older
respondents aged 75 and above reporting a mean score of 3.48 (95%CI:
3.45–3.51) as compared to younger respondents aged 15–24 (2.86,
95%CI: 2.83–2.89). There was little variation in the mean score for
continuity of care with respect to sex, however, sex CIs do not overlap.
Statistically significant variability of continuity of care was also observed
for marital status and ethnicity.
Income was negatively associated with continuity of care
score. For example, those in the lowest income tertile had a mean continuity of
care score of 3.23 (95%CI: 3.21–3.25) and those in the highest income
tertile had a mean continuity of care score of 3.02 (95%CI: 3.00–3.04).
Statistically significant variability of continuity of care was observed for
labour force status: those not working had a higher mean continuity of score
(3.24, 95%CI: 3.22–3.26) as compared to those who were working (3.02,
95%CI: 3.01–3.03). There was little evidence for significant variation of
continuity of care with NZDep, NZiDep, or education. In contrast, significant
variability was observed for the smoking covariate.
Table 1.
Demographic, socioeconomic and health characteristics of study population by
mean continuity of care score: SoFIE-Health,
2004–051
Note:
1Total N may not sum up to 16630
because of random rounding.
Ex-smokers had the highest mean continuity of care scores
(3.17, 95%CI: 3.15–3.19) while current smokers had the lowest score (3.06,
95%CI: 3.04–3.08). Mean scores for continuity of care do not differ with
the levels of psychological distress (Kessler-10). However, there is an
increasing trend in continuity of care with increasing numbers of co-morbid
diseases. Those reporting no co-morbid conditions had a lower mean continuity of
care score (2.99, 95%CI: 2.97–3.01) than those reporting 2 or more
co-morbid conditions (3.35, 95%CI: 3.32–3.38).
To check the effect of controlling simultaneously for all
covariates, we performed an OLS regression analysis. Table 2 presents results
from the OLS regression analysis, in which different predictors are regressed on
continuity of care simultaneously controlling for demographic, socioeconomic,
health behaviour and health factors. We sequentially added demographic,
socioeconomic and health and health behaviour variables to the model but, for
brevity only, the results from the final model are presented.
Results from this analysis show that the estimated
coefficients for all the demographic factors were significant. Individual
coefficient estimates suggested that age was significantly associated with an
increase in continuity of care, while male sex and never married were associated
with a reduction in continuity of care. Continuity of care increased by 0.19 and
0.06 points for the Pacific and Asian ethnicities, by 0.09 points for those in
the lowest income tertile, by 0.06 for those not working, and by 0.14 points for
those with 1 or more co-morbid conditions compared with their respective
reference categories (see Table 2). However, continuity of care decreased with
an increase in individual deprivation characteristics. It is important to note
that OLS results are consistent with the bivariate results shown in Table
1.
Table 2. Estimates from OLS for continuity of
care, adjusting for effects of demographic, socioeconomic, health behaviour and
health variables: SoFIE-Health,
2004-051
Note:
1SoFIE= Survey of Family, Income and
Employment
DiscussionOverall, our data provide some support for the hypothesis
that people with high health needs have higher mean continuity of care score
(e.g., the elderly, Pacific and Asian ethnic groups, those with low incomes, and
those with one or more chronic conditions). The finding that older people had a
higher continuity of care mean score probably reflects an increase in chronic
conditions and other morbidities with age.
Although this research raises several important findings
related to continuity of care to primary health care using national survey data,
there are several limitations to this study that must be considered when
interpreting the results. First, this study reports cross-sectional associations
which prohibit drawing causal inferences. Follow up data (Wave 5) may allow more
progress in deducing causal relations. Second, given that continuity of care was
measured on self-reported data not confirmed by a physician/administrator, our
estimates may be subject to reporting error and recall bias not accounted for by
statistical adjustments. Third, Asian and Pacific ethnicity did not take into
account cultural variations within these large, heterogeneous groups.
Another limitation is attrition in the data. In Wave 3 of
the SoFIE study, 83% of the original sample members were re-interviewed,
18 which combined with the household response
rate at Wave 1 of 77% gives an estimated effective response rate of 64%. While
attrition within the SoFIE study is low compared with other population-based
longitudinal panel surveys, 19 20 selection
bias might arise in our analyses if individuals drop out of the survey in a
non-random manner (i.e., the more unhealthy may be more likely to not
participate in follow-up years). It is not possible to estimate whether such
bias occurred. This would require selective attrition within strata of key
covariates, which seems unlikely.
It is also important to note that the coefficient of
determination for the continuity of care estimator was small,
R2=0.10, indicating the model did not account
for much of the observed variation in continuity of care. Hence, the model has
little predictive power. The same caveats apply to models applied separately to
each of the questions that comprise the continuity of care measure. This may be
due to the overall high and relatively invariant levels of continuity of care in
New Zealand.
Despite these limitations, the results presented here are
important in several ways. This study uses a large, original, national survey in
creating a continuity of care index at individual patient level. Few previous
studies in New Zealand have focussed on primary care attributes at an individual
patient level. This may, in part, be due to the challenge of collecting
information at the individual level about aspects of primary care or the
inability of consumers to be valid judges of primary care quality.
14
Our results have important implications for health care
policy, especially as cost containment and cost effectiveness has become
increasingly important. Continuity of care has been found to be associated with
lowering health care costs 3 among patients by
decreasing use of emergency services 8 9 and
hospitalisation 10 11 and also because primary
care physicians provide care that is less costly than secondary care.
21 Thus it can be argued that encouraging and
motivating patients to form a consistent relationship with their PCP may result
in reducing costs of health care.
The current Primary Health Care Strategy requires
individuals to be enrolled/registered with a Primary Health Organisation /
General Practitioner (GP) in order be eligible for lower GP consultation fees.
In 2003-04, nearly 92% of the NZ adult population was affiliated with a PCP.
23 High affiliation with a PCP may lead to high
continuity of care. The authors propose that continued incentives to develop and
sustain affiliation with a PCP and continuity of care are important for
maintaining the quality and cost-effectiveness of primary health care.
Competing interests: None.
Statistics New Zealand Security
Statement: Access to the data used in this study was provided by
Statistics New Zealand in a secure environment designed to give effect to the
confidentiality provisions of the Statistics Act, 1975. The results in this
study and any errors contained therein are those of the authors, not Statistics
New Zealand.
Disclaimer: Opinions expressed in this
report are those of the authors only and do not necessarily represent the views
of the peer reviewers or the University of Otago.
Author information: Santosh Jatrana, Senior
Research Fellow; Peter Crampton, Dean and Head of Campus; Ken Richardson, Senior
Research Fellow; Department of Public Health, University of Otago,
Wellington
Acknowledgements: SoFIE-Health is primarily
funded by the Health Research Council of New Zealand as part of the University
of Otago’s Health Inequalities Research Programme. Establishment funding
was also received from the University of Otago, Accident Compensation
Corporation of New Zealand (ACC), and the Alcohol Liquor Advisory Council
(ALAC). Comments on this paper were received from Tony Blakely. We are grateful
for the contribution of Ken Richardson and Kristie Carter in preparing the data
set. Lastly we thank the anonymous peer reviewers for their insightful comments
on this paper.
Correspondence: Santosh Jatrana, Department
of Public Health, University of Otago, Wellington, PO Box 7343, Wellington, New
Zealand. Fax: +64 (0)4 3895319; email: santosh.jatrana@otago.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|