October 2006

You are currently browsing the monthly archive for October 2006.

The effect of financial incentives on gatekeeping doctors

October 31, 2006 in Academic Articles, International Health Care Systems, Physician Compensation | Permalink

In 1991 in the UK, the British began allowing general practitioners (GPs) to participate in a fundholding scheme. The fundholding program would reimburse GPs if the amount of chargeable elective secondary care procedures was below their budget and financially penalize the GPs if the amount of secondary care which their patients received exceeded their budget. By 1997, over half of the British population went to a GP who participated in a fundholding system. Later in 1997, the Labor party won the election and the more liberal parliament decided to outlaw the fundholding system.

If one were to analyze the decision to participate in the fundholding scheme in the early 1990s, this analysis is prone to selection bias–since physicians who treat patients less aggressively are more likely to participate–and thus studies such as Croxson et al. (JPubE 2001) are likely to have inaccurate results. On the other hand, the change to non-fundholding system is an exogenous shock caused by a government mandate. Dusheiko, Gravelle, Jacobs and Smith (JHE 2006) use this natural experiment to perform various difference in difference (DD) estimation strategies to determine the effect of financial incentives on gatekeeping doctors.

Their data comes from admission information from over 7000 English practices in the years between 1997 and 2001. The results from a DD regression estimates suggest the “effect of removing the financial incentives of holding a budget was to increase chargeable elective admissions amongst the practices that elected to become fundholders by 3.5-5.1%.” Looking at the conditional means in 1997 (the year before the abolition of fundholding), the authors find that fundholders had a 7.3% lower admission rate than non-fundholders. The paper finds that approximately 60% of this difference was due to financial incentives and 40% was due to unobserved differences between the characteristics of fundholding and non-fundholding practices.

Some procedures (such as ER visits) were not counted against the GPs fundholding budget. One would expect that the amount of non-chargeable admissions would not change after the 1998 abolition, however, it might be possible for GPs to send a patient for a secondary procedure coded as non-chargeable even though it should have been coded as a chargeable admission (similar to DRG creep in the U.S.). The authors claim to find no evidence of this; the amount of non-chargeable admissions does not vary significantly before and after the abolition. Dusheiko and colleagues also use this fact to perform a difference in difference in difference (DDD) estimation and find similar results to the DD regression.

A significant confounding factor in this data is that patients in the UK often have to wait a significant amount of time to receive treatment. In fact, the average wait time in this data set was 100 days. Thus, one may worry that admissions in the current year are based on decisions made in the prior year.

Dusheiko; Gravelle; Jacobs; Smith (2006) “The effect of financial incentives on gatekeeping doctors: Evidence from a natural experiment” Journal of Health Economics, vol 25, pp. 449-478.

HMO quality: Separating perception from reality

October 31, 2006 in Health Insurance, Managed Care | Permalink

It has been shown in various studies and opinion polls that consumers generally believe that HMOs provide an inferior level of care than non-HMO plans. This is true even when more objective measures of medical service quality are taken into account. Why is HMO satisfaction so low?

A study by Reschovsky, et al. (2002) claims that people who are dissatisfied with their medical care are more likely to report that they have an HMO. The authors use the Community Tracking Study. The CTS asks survey respondents what type of insurance they have. Later, they contact the individual’s insurance company in order to glean the details (copayment and coinsurance rates, deductibles, referral requirements, insurance type, etc.) of the person’s true insurance coverage.

As an outcome variable, the authors looked at various patient satisfaction measures (e.g.: the level of trust they have with their current physician, their satisfaction with their last doctor’s visit). When they compare people who have an HMO and correctly report this, with those who have an HMO but report they have non-HMO coverage, the authors find that those with the correct reporting have lower satisfaction scores. On the other hand, comparing people with a non-HMO insurance who correctly report this with those who have a non-HMO but report having an HMO, they find the people incorrectly reporting that they have an HMO had lower satisfaction scores. Below is the results for the dependent variable of “percent dissatisfied with their health care in general.”

  HMO-Actual Non-HMO Actual
HMO Reported
 9.6% 10.1%
non-HMO Reported
 6.3% 7.3%

Thus, they conclude that reporting of HMO coverage may be negatively correlated with actual satisfaction and may not accurately reflect the survey respondant’s true coverage. As the title indicates: “It’s not whether you are in an HMO but whether you think you are”

Reschovsky; Hargraves; Smith (2002) “Consumer Beliefs and Health Plan Performance: It’s Not Whether You Are in an HMO But Whether You Think You Are” Journal of Health Politics, Policy and Law, Vol 27, No. 3 pp.353-377.

Hausman Endogeneity Test

October 30, 2006 in Econometrics | Permalink

Ordinary Least Squares 

If you have studied basic statistics, its likely that you have come across the ordinary least squares (OLS) estimation technique.  OLS attempts to minimize the squared distance between dependent variables (‘y‘) and the a linear prediction of y (y_hat=xβ).  The parameter vector ‘β_ols‘ minimizes this distance.  The most important assumption in order for β to reflect to true parameters in the population is for the regressors to be uncorrelated with the error terms (cov(x,e)=0).  Sometimes this is not the case.  The assumption fails if:

  1. There are omitted variables which are correlated with the regressors (x)
  2. We have a system of simultaneous equations.
  3. There is an errors in variables problem
  4. The system has a lagged dependent variable with a serially correlated disturbance

Instrumental Variables

One solution to these problems is to use an Instrumental Variables (IV) technique.  (Click here for an explanation of IV).  A question remains as to when OLS is appropriate and when IV is best.  OLS will generally give smaller standard errors (and thus is more precise) and is to be preferred when the β_OLS parameters are unbiased.  

Hausman Endogeneity Test

To test whether the IV or OLS regression technique is best, one can use the Hausman endogeneity test.  Let us try to estimate the following equation:

  • (1) y1 = x1*δ + y2*α + e

Let the vector z=(x1,x2) be the set of all exogenous variables.  The vector x1 is the set of regressors and x2 are our instruments.  Since z is exogenous, we know E(z′*e)=0.  The variable y2, we believe to be endogenous. 

One example of an endogenous y2 would be a wage equation where y1 is the individual’s wage and y2 is the number of hours worked.  We would think that full time workers would earn more than part time workers so hours would affect wage.  On the other hand, when a worker’s wage is higher (assuming leisure is a normal good) one would expect the individual to work more hours.  In this example we have dual causation.

To conduct the Hausman test, we first find the linear projection of y2 on z using OLS.

  • (2) y2 = z*π + v

Since the error term from the first equation (‘e‘) is uncorrelated with z by assumption, then y2 is endogenous if and only if E(v*e)≠0.  We can test whether the structural error ‘e‘ is correlated with the reduced form error (‘v‘) using the following equation:

  • (3) e = Ï?*v + u

If we plug equation 3 into equation 1 and we have:

  • (4) y1 = x1*δ + y2*α + Ï?*v + u

In empirical data, however, ’v‘ is not observed.  Nevertheless, we can estimate ‘v_hat‘ by taking the saved residuals from our OLS regression in equation 2 and plugging these numbers into equation 4 for ‘v’.  The final equation is:

  • (5) y1 = x1*δ + y2*α + Ï?*(v_hat) + u

We can now consistently estimate δ,α, and Ï? using OLS.  Using the usual OLS t-statistic, we can test the null hypothesis that Ï?=0.  If we accept the null, then there is no endogeneity problem and one should use an OLS estimation strategy.  If Ï?≠0, then the instrumental variables technique is best.  One can also use a heteroskedasticity-robust t-statistic for testing Ï? if one suspects heteroskedasticity. 

A similar set of procedures can be extended to the case where y2 is a vector.  Instead of an t-test on the residual ‘v_hat‘, in the vector case we would have to preform an F-test (Ï?=0) on a vector of residuals ‘v_hat‘.  To see how to preform Hausman tests in the Stata statistical package, look at this paper by Baum, et al.

Summary

  1. Preform first stage regression of the endogenous variable (y2) on z.
  2. Calculate the residuals from this equation and include them as an additional regressor in the original estimation equation.
  3. Run OLS on this new equation and preform a t-test for the coefficient on the first stage residuals.
  4. If one accepts the null hypothesis, then there is no endogeneity problem and OLS should be used.  If one rejects the null hypothesis, then endogeneity is a problem and one should use an IV estimation strategy.

References

Wooldridge, Jeffrey; Econometric Analysis of Cross Section and Panel Data, MIT Press, London, (c) 2002, pp. 118-122.  

Decrease price…increase supply?

October 27, 2006 in Medicaid/Medicare, Supply of Medical Services | Permalink

In a typical market, an increase in the consumers’ willingness to pay will increase price and increase quantity (see graph).  On the other hand, a decrease in willingness to pay will decrease price and decrease quantity. 

This axiom of economics does not hold in the health care market; at least not according to a 1998 HCFA White paper to Richard Foster.  The paper found that when Medicare decides to reduce its fees, the quantity of medical services supplied by physicians actually increases.  In fact, a Medicare price decrease led to increased medical service volume and intensity by 31% (significant at the 5% level).  A Medicare price increase also increased volume and intensity but the results were not statistically significant. 

Other papers have found similar results:

  • Yip (1994) found that after Medicare reduced the price for coronary artery bypass graftings (CABGs) in New York and Washington, there was a large and statistically significant increase in the volume and intensity of CABGs.
  • Christensen (1992) looked at the state of Colorado in the 1970s.  He found that a one half of a Medicare price decrease was offset by increased volume of medical services and one third of the price decrease was offset by increased intensity of medical services.
  • Nguyen and Derrick (1997) estimated a behavioral response which was statistically significant only among physicians whose practices received a Medicare price reduction.  The magnitude of the response was 40% for these firms.

Why is this occurring?

  1. First, patients often do not know what type of care they require so they physicians can suggest treatments which may be unnecessary.
  2. Since patients bear little cost of these procedures, their price sensitivity is low relative to the total cost of a procedure.  It is also possible that when Medicare prices decrease, the amount of the coinsurance paid by patients needed to cover a procedure goes down; this demand change may have some influence on the increase in quantity.
  3. Uncertainties in the practice of medicine allow for a variety of practice styles, so even peer review or a physician’s behavior may not be appropriate. 

Academic Integrity

October 26, 2006 in Economics - General | Permalink

In order to prevent fraud, such as in the case of Eric Poehlman, economists use a peer-review system consisting of referee reports.  The referee reports evaluate the merit of a paper and offer suggestions of how to improve the article.  Tyler Cowen of Marginal Revolution gives some good tips on the best way to write referee reports.  Kwan Choi also has a concise article giving pointers of how to write a good referee report. 

Those annoying HMO patients…

October 25, 2006 in Physician Compensation | Permalink

Why do HMO patients receive less care than fee-for-service patients?  Could it be that HMO patients are healthier (adverse selection) or that FFS compensation leads to increased demand for medical services (moral hazard)?  A paper by Shen, et al. (2004) finds that one reason could be that physician compensation could affect a doctor’s desire to perform discretionary care. 

Methodology

The authors of the study sent out a survey to a nationally representative group of family practice physicians.  Each physician was presented with four vignettes covering a spectrum of a family practitioner’s practice.  The vignettes were:

  1. outpatient drug prescribing,
    • a 50-year-old man with arthritis for which he takes over-the-counter medication but would prefer a more costly prescription drug because it “seems strongerâ€? and is more convenient. He requests the prescription drug”
  2. diagnostic test ordering,
    • a healthy 28-year-old woman in early pregnancy desires a fetal ultrasound “to make sure things are OK … just like I had with my last child.â€?
  3. specialist referral,
    • a 5-year-old boy is brought in by his mother for a dog bite on his cheek. The mother reminds the physician that a plastic surgeon was asked to close a facial laceration on her daughter a few years earlier.
  4. the management of end-stage heart failure
    • a 61-year-old man with congestive heart failure whose only remaining therapy is a heart transplant

Physicians were randomly assigned patient vignettes in which the individuals were FFS or capitation patients.  The goal of the study was to find whether physicians treated FFS patients differently from capitation patients, as well as to investigate whether or not the doctors felt more “bothered” performing services for one of the groups.

Results

After surveying approximately 800 family practice physicians, the authors found that doctors were more likely to provide discretionary care in the 1) outpatient drug prescribing, 2) diagnostic test ordering, and the 3) specialist referral vignettes.  Using a logit regression the authors found the results to be significant below the 1% level.  For vignette 4, the management of end-stage heart failure, there was no statistically significant difference in discretionary care between the two groups.  It was also interesting to find that the physicians felt more ‘bothered’ when treating capitation patients in all four scenarios.

Although survey data does not demonstrate what physicians will do in practice, it does seem that homo economicus does drive-at least in part-family practice physician decision-making.

Shen; Andersen,;Brook; Kominski; Albert; Wenger (2004) “The effects of payment method on clinical decision-making“ Medical Care, vol 42(3), pp. 297-302.

 

I love my doctor

October 24, 2006 in Medical Studies | Permalink

People love their doctors.  Study after study has shown this to be true.  Typically 75%-95% of patients surveyed claim that they completely or mostly trust their physician.  Does this finding hold even when doctors are compensated through a capitation method and have a financial incentive to withhold care?  A study by Kao, et al. (1998) shows that even capitation patients largely trust their doctor.  The authors find 94% of FFS indemnity patients, 85% of managed FFS patients, 83% of salary patient patients and 77% of capitated patients trust their doctor. 

A study by Hall et al. used a randomized trial where half of HMO patients were explicitly informed that their doctors were compensated on a capitation basis and had a financial incentive to withhold services.  These disclosures were found to have no negative effects on the trust of either physicians or insurers.  The authors conclude that “once patients have formed opinions of their physicians based on actual experience with them, information about payment methods has little or no effect on their opinions, possibly because of the cognitive dissonance that otherwise would arise.”  It would be interesting to examine whether a doctor’s actual proficiency or the doctor’s personality mattered more to the patient in terms of their level of trust. 

Hall; Dugan; Balkrishnan; Bradley; (2002) “How disclosing HMO physician incentives affects trust” Health Affairs, March/April, pp. 197-206.

Kao; Green; Zaslavsky; Koplan; Cleary; (1998) “The relationship between method of physician payment and patient trust” JAMA, Vol 280, No. 19, pp. 1708-1714.

An Unwelcome Discovery

October 24, 2006 in Current Events | Permalink

While I was reading the The New York Times Magazine, I came across a very interesting article regarding scientific integrity.  The article (“An unwelcome discovery“) documents how a Eric Poehlman, a faculty member at the University of Vermont, had fabricated ten years worth of data.  This is a serious problem .  Sally Jean Rockey of the N.I.H. explains the gravity of the situation best:

“Science is incremental,â€? she said, explaining that most scientific advances build on what came before. “When there’s a break in the chain, all the links that follow that break can be compromised.â€? Moreover, she said, fraud as extensive as Poehlman’s would inevitably lead to further erosion of the public’s trust in science. 

Dr. Pohelman was sentenced to one year in prison and two years of probation and is banned from receiving any public research dollars for life.  Some of Poehlman’s fraudulent activities include:

  • Changing results: Poehlman hypothesized low-density lipoprotein (LDL) would increase as individuals age while high-density lipoprotein (HDL) would decrease over time.  When a longitudinal study which he set up did not produce the desired results Poehlman changed the number manually.
  • ‘Inventing’ data: Poehlman was most noted for a longitudinal study on menopause.  Yet these results were also completely fabricated.  Poehlman had tested only 2 women, not 35 as he had claimed.   

The fabrication of this data lead to Poehlman receiving millions of dollars in NIH grant money.  This money comes from working men and women who had entrusted scientists to work to discover the next cures to make their lives–or their children or grandchildren’s lives–better.  In the case of Poehlman, this trust was not warranted.  Integrity is needed from today’s researchers.  Without integrity, the steady march of scientific progress made over the centuries could come to a grinding halt. 

Diagnose your disease

October 23, 2006 in Medical Studies | Permalink

Are you feeling ill or experiencing worrying symptoms?  Interested in looking for information regarding a disease?  If so, you can find answers to these and other medical questions at Wrong Diagnosis.  This site uses reliable sources to create it medical database.  These sources include:

  • the Center for Disease Control and Prevention (CDC),
  • the National Institute of Health (NIH),
  • the Food and Drug Administration (FDA),
  • the Department of Health and Human Services (DHHS)
  • the U.S. Surgeon General
  • National Vital Statistics Reports (NVSR)
  • A UK Diseases Database, (Medical Object Orientated Software)
  • the Australian Victoria state study (DHS-VIC)

 

Media that Matters – Women in the Democratic Republic of Congo

October 20, 2006 in Health Care in Developing Nations | Permalink

Media that Matters is an organization dedicated to producing films portraying controversial issues relevant to today’s society and encouraging viewers to work towards social change.  While the organization is heavily left leaning politically, they do have a wide variety of health-related films (see health/health advocacy section).   I did come across an interesting short film regarding women in the Congo (map) titled In Transit.  Below, you can find an excerpt from a discussion of In Transit by the film’s directors.  The directors plan to make a full length version of the movie in the future.

“The Goma Film Project came about when Doctors on Call for Service (DOCS—now called HEAL Africa), a teaching hospital and nonprofit organization located in Goma, Democratic Republic of Congo (DRC), enlisted filmmakers Louis Abelman, Bent-Jorgen Perlmutt, and Nelson Walker III to produce assorted informational videos. The first of our assignments was to create a small series of surgical training videos documenting vesico-vaginal fistula (VVF) repair, the most common procedure performed at DOCS on any given day.

A vesico-vaginal fistula is a tear in the vaginal wall that opens into the urethra. In the first world, fistulas are rare; in most parts of the third world, they are caused by complications during childbirth. In the eastern Congo, they are primarily caused by an epidemic of violent rape and torture taking place in the war-ravaged countryside. Marauding armies and militias, some responsible for the Rwandan genocide of 1994, continue to stalk the forests and villages of the region. Women and girls of multiple tribal and ethnic backgrounds have increasingly become victims of plunder, rape, and slavery by the many armed groups involved in the ongoing conflict. At least 40,000 rape cases have been documented in the past several years, with some estimates of up to a million additional undocumented cases. DOCS/HEAL Africa is the only hospital in the region that offers a comprehensive rehabilitation program for women who have acquired fistula from rape.”

« Older entries