Investigating the Effects of Private Public Partnerships on Delivery of Public Health Services

Document Sample
Investigating the Effects of Private Public Partnerships on Delivery of Public Health Services Powered By Docstoc
					    Investigating the effects of
  private-public partnerships on
delivery of public health services 1

                    Sergey Sotnikov, PhD
                       Toby Merlin, MD
        Division of Partnerships and Strategic Alliances
             National Center for Health Marketing
          Centers for Disease Control and Prevention
                        ann0@cdc.gov
 1The findings and conclusions in this presentation have not been formally
disseminated by CDC and should not be construed to represent any agency
                         determination or policy.
        RESEARCH OBJECTIVES
   The recent report from Institute of Medicine
    (IOM), “The Future of the Public’s Health in the
    21st Century,” promotes partnerships as an
    important way to improve public health.
   However, the evidence of the positive effects of
    partnerships on measurable public health
    outcomes is scarce.
                 STUDY GOAL
   This study intends to provide quantitative
    evaluation of the effects of public-private
    partnerships on the number and the types of
    services provided by the local health departments
    (LHDs).
          POPULATION STUDIED
   The main sources of data on LHD characteristics,
    partnerships, and types of 36 services provided by LHD
    is the 1996 NACCHO survey of 2793 LHD.
   The 36 service types include: immunizations, animal
    control, behavioral health, child health, chronic disease
    prevention, communicable disease control, community
    assessment, dental heath, environmental health, family
    planning, HIV/AIDS, STD, TB, health education, home
    care, injury control, inspections, lab services, mental,
    obstetrical, prenatal and maternal health, occupational
    health and safety, primary care, programs for homeless,
    school based health, substance abuse, tobacco
    prevention, and veterinarian services.
                STUDY DESIGN
   Private-public partnerships can be viewed as
    organizational interventions aimed at improving the
    efficiency of public health service delivery.
   The gold standard of study design for quantification of
    the effects of interventions is randomized control trial
    (RCT). However, RCT is not feasible in the case of
    partnerships because partnering is a matter of choice,
    and, thus, random assignment of LHD into partner and
    non-partner groups is not possible.
   Partnership effects estimated without accounting for
    non-random selection will be biased.
         Characteristics of local health departments
     according to their partnership status with businesses
                                do not have    do have      Difference
                                partnership   partnership         (%)
                                    with          with
                                businesses    businesses
Expenditures (dollars)          3917637       8028140          105
Number of employees                66           124            87
Number of full time employees      49           103            109
Expenditures per capita
  (dollars)                        33           38             13
Employees per 100000
  population served                91           102            12
Population served                71807        137053           91
Number of services provided        18           22             23
                  1.1
                        1.2
                              1.3
                                    1.4
                                          1.5
                                                1.6
                                                      1.7
                                                            1.8
                                                                  1.9




              1
                                                                        2
    ilrest
     aiflu
chimmun
aipneumo
  aihepb
  animal
   ilrecw
    tbtest
    ehall
  ehrecw
  comout
      ilall
  episurv
 healthed
   ilpubw
   ilprivw
     labs
      tbtx
 famplan
 ehsurfw
    stdtc
ehemerg
 cdcansc
     stdtx
     hivct
    cdall
ehhazmat
                                                                                        according to business partnership status




  cdhbptx
   cdcdtx
  tobacco
 cddiabtx
schoolhe
                                                                            Observed Differences in Provision of Public Health Services for LHD




cdglausc
   dental
 schoolcl
  occsafe
 primcare
 cdglautx
        STUDY DESIGN - continued
   We use propensity-scores-matching methodology to
    conduct quasi-experimental assignment of LHDs into
    comparable pairs of cases and controls.
   Pscore routine in STATA9 is used to estimate a Probit
    model of business partners’ choice as a function of
    observable characteristics (LHD expenditures, number of
    customers in jurisdiction, number of employees, the size of
    the jurisdiction).
   Predicted probabilities of having a business partner are
    used to match LHDs with and without business partners by
    nearest-neighbor-matching method.
        STUDY DESIGN - continued
   The effects of LHD private partnerships are
    estimated by calculating the difference in outcome
    variables for each pair and taking a mean of the
    differences.
   The outcomes are dichotomous variables
    indicating if any of the 36 services are provided by
    LHD and also the total number of services
    provided.
         PRINCIPAL FINDINGS
   LHDs that partner with businesses are more likely than
    LHDs that do not have business partners to deliver the
    following services:
    injury prevention (11.3% point difference, t=3.65),
    school health programs (7.6%, t=2.70),
    tobacco control (6.4%, t=2.38), and
    community assessment (5.3%, t=2.32).
   LHDs that cooperate with businesses tend to provide
    less obstetrical care (-6.2%, t=-2.00).
   No statistically significant effects (at 0.05 level) of public-
    private partnerships were observed for all other services
    as well as for the total number of services delivered.
          Percent of local health departments providing
                             services

100


 80


 60


 40


 20
                          11            8               6               5
  0
            -6
      Obstetrical     Injury     School health    Tobacco         Community
        care        prevention     programs        control        assessment
-20

                    LHD do have partnership with businesses
                    LHD do not have partnership with businesses
                    Difference (% points)
               CONCLUSIONS
   This study suggests that LHD partnerships with
    businesses may have beneficial effects on
    delivery of injury prevention, school health,
    tobacco control, and community assessment
    programs.
     IMPLICATIONS FOR POLICY,
     DELIVERY AND PRACTICE
   Promoting LHD partnerships with businesses
    may have heterogeneous effects on delivery of
    public health services, i.e. delivery of some
    services may improve more than others
                  Percent of local health departments that partner with

                   100
                     90
                     80
                     70
                     60
                     50
                     40
                     30
                     20
                     10
                      0
                           Community
                                                                Physician                   Faith
                             health    Hospitals   Nonprofits               Businesses
                                                                practices                communities
                             centers
1996                          39          67          64           59          44            37
2000                          54          90          74           90          74            80
2005                          57          88          89           92          80            82
2006 compared to 1996 (%      18          21          25           33          36            45
points)
      Contact Information

            Sergey Sotnikov, PhD
Division of Partnerships and Strategic Alliances
     National Center for Health Marketing
  Centers for Disease Control and Prevention
                ann0@cdc.gov

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:64
posted:4/12/2008
language:English
pages:15