phc estimation human resource needs

S30 Revue canadienne de la pratique et de la recherche en diététique – Supplément, automne 2006

AbstractPurpose: Information on human resources and costs is

needed to plan for the addition of registered dietitian (RD)services to new models of primary health care (PHC). Estimateswere developed, based on an analysis of an enhanced RD modelof counselling and health promotion services in three OntarioFamily Health Networks (FHNs).

Methods: Both direct and indirect costs were averaged overthe three FHNs. Costs and RD activities were tracked through-out 2005. The FHN staff completed two questionnaires address-ing communication, case management, and satisfaction with RDservices.

Results: Actual and reported case management indicatedthat an estimated 1.3% to 2.4% of the 60,000 enrolled patientsmay require individual nutrition counselling in a year. If onefull-time equivalent (FTE) RD can manage 380 new referrals,then one FTE RD is needed per 15,800 to 29,000 patients. Theestimated direct costs of adding one FTE RD (including expensesand fixed costs) is $78,169 to $80,169, when the RD is an inde-pendent contractor.

Conclusions: Additional studies are needed to develop betterestimates of human resource needs and costs of interdisciplinarynutrition services in all PHC settings. These estimates should bebased on population characteristics and direct and indirect costsfor all models of nutrition services in PHC settings.(Can J Diet Prac Res 2006;67 Suppl:S30-S38)

RésuméObjectif. Nous avons besoin d’information sur les ressources

humaines et les coûts pour planifier l’insertion des services dediététistes professionnelles (DP) dans les nouveaux modèles desoins de santé primaires (SSP). Des estimations ont été établies,basées sur une analyse d’un modèle amélioré de services decounselling et de promotion de la santé dispensés par des DPdans trois réseaux de santé familiale de l’Ontario (RSF).

Méthodes. On a calculé les coûts moyens directs et indirectsdans les trois RSF et relevé les coûts et activités des DP pen-dant toute l’année 2005. Le personnel des RSF a rempli deuxquestionnaires sur la communication, la gestion des cas et lasatisfaction vis-à-vis des services de DP.

Résultats. Selon la gestion des cas réels et rapportés, de 1,3 à2,4 % des 60 000 patients inscrits nécessiteraient un counsellingnutritionnel individuel en une année. Si un équivalent tempsplein (ETP) de DP peut traiter 380 nouveaux patients, on aurabesoin d’un ETP par 15 800 à 29 000 patients. Coût estimé del’ajout d’un ETP de DP (incluant dépenses et coûts fixes) : entre78 169 $ et 80 169 $, lorsque la DP est une entrepreneureindépendante.

Conclusions. Des études supplémentaires sont nécessairespour obtenir de meilleures estimations des besoins en ressourceshumaines et des coûts liés aux services nutritionnels interdisci-plinaires dans tous les milieux de SSP. Ces estimations devraientêtre basées sur les caractéristiques de la population et sur lescoûts directs et indirects relatifs à tous les modèles de servicesnutritionnels en milieu de SSP.(Rev can prat rech diétét 2006;67 Suppl:S30-S38)

Estimation of Human Resource Needs And Cost of Adding Registered Dietitians

To Primary Care NetworksJULIA WITT, PhD, Melbourne Institute of Applied Economic and Social Research, The University of Melbourne, Melbourne, Australia;

PAULA BRAUER, PhD, RD, Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, ON; LINDA DIETRICH, MEd, RD, Dietitians of Canada, Toronto, ON;

BRIDGET DAVIDSON, MHSc, RD, Nutrition and Research Consulting, Kitchener, ON

STEERING COMMITTEEPaula Brauer, PhD, RD, Linda Dietrich, MEd, RD, Bridget Davidson, MHSc, RD, John Krauser,

Primary Health Care Team, Ministry of Health and Long-Term Care, Toronto, ON, Karen Parsons, Primary Health Care Team, Ministry of Health and Long-Term Care, Kingston, ON

Successful integration of registered dietitians (RDs)into primary health care (PHC) organizations will beachieved by creating feasible, effective services that meetthe needs of health care providers and patients, at a pricethat the Canadian health care system can afford. Whenthey are considering adding an RD to the team, serviceplanners need information about the projected direct costs,including salary and expenses, as well as the indirect costs tothe rest of the organization. Therefore, as part of an Ontario

Primary Health Care Transition Fund demonstrationproject, preliminary human resources and cost estimateswere developed, based on data collected when three RDswere placed in three Family Health Networks (FHNs).

Family Health Networks are an Ontario model of primarycare with three or more family physicians (FPs) workingwith other selected health care professionals to providePHC services to enrolled patients. They are funded in ablended funding model that includes capitation payments

INTRODUCTION

RESEARCH

RECHERCHE

3006055DC_Research BrauerCost_S30_S38 8/9/06 1:44 PM Page 30

RESEARCH

RECHERCHE

Canadian Journal of Dietetic Practice and Research – Supplement, Fall 2006 S31

(population-based funding of health care services) withincentives for additional specific preventive health careactivities. These FPs often work in separate offices. Theyhave received support to implement electronic healthrecords (EHRs). Family Health Networks have not had RDsin the past.

METHODSContext: dietitian practice in each FHN

The Steering Committee reasoned that better estimatesof average human resource needs and costs would be devel-oped for planning purposes if geographically diverse FHNswere represented. In April 2004, a request for proposals wassent to all FHNs and primary caremodels that met specific criteria,including five or more physicians,diverse locations in urban, rural, andnorthern settings, development ofEHRs already underway, and no pre-vious RD services. The three FHNs(Parry Sound, Kingston, and Strat-ford, Ontario) were chosen in May2004 from eight submissions. In July, Dietitians of Canada(DC) hired three experienced RDs as independentcontractors to work at each of the sites. The RDs receivedorientation and training in August, and began working inFHN offices in September 2004 as directed by staff ateach FHN. The RDs were contracted to work seven hoursa day for three or four days a week, depending on availableproject funding.

Each FHN organized its RD services differently. In oneFHN with three separate offices, the RD was located in oneoffice, had to arrange her own appointments, and hired astudent to call patients for initial appointments. This RDbooked all her own follow-up appointments. In the secondFHN with 13 offices, the RD had an office in one location,and a receptionist or nurse booked the appointments. Inthe third FHN of three offices, the RD provided nutritioncounselling at each site on given days of the week, and ini-tially a receptionist booked the nutrition appointments.(Halfway through the project, the RD was required to takeover the booking of all initial and follow-up appointments.)This RD also did not have an office, and carried all herresources, including her computer and nutrition patientcharts, to each site. All FHNs were in the process of imple-menting EHRs over the course of the project, but only one FHNhad a fully functioning EHR system by project completion.

The RDs implemented a practice model that was broadlybased on previous role documents (1,2), with a main focuson nutrition counselling and a secondary mandate to develophealth promotion and disease prevention programming.Elements of “enhanced” practice were implemented (3,4).These included computerized diet record analysis, a coun-selling process using the PRECEDE-PROCEED model (5),and assessment of blood pressure, waist circumference, andhealth-related quality of life (SF-36) (6-9) with a minimumof two client visits (a baseline assessment and three-monthfollow-up). In addition, a portion of each baseline appoint-ment was devoted to explaining an RD clinical evaluation

study and to soliciting possible participation. Otherwise,the RD and her clients developed individualized care plans,based on mutually agreed-upon goals. Work for the projectwas completed in March 2006.

Questionnaire and data collection form development All questionnaires and data collection forms were

adapted from other sources. The physicians’ managementform, clinical monitoring forms, and workload measure-ment categories were adapted from the Hamilton HealthService Organization (HSO) Mental Health and NutritionProgram instruments (Anne Marie Crustolo, personalcommunication 6 Feb 2004), to ensure that comparable

data would be collected.A change-in-routine questionnaire

was developed to identify any majorworkflow disruptions that the RD mayhave caused FHN staff (10,11). Eachstaff member received a letter explain-ing the study, the questionnaire, anda stamped envelope addressed to theUniversity of Guelph. No identifying

information was collected. The person responsible foradministering the survey at each FHN sent an e-mailreminder to staff halfway through the data collectionperiod.

All questionnaires and data collection forms were pilottested or reviewed by the RDs and the lead physicians. A sur-vey methodologist also reviewed all forms and questionnaires.In addition, 14 RDs from a primary health care advocacygroup reviewed the change-in-routine questionnaire.

Overall data collection planThe three RDs and the project coordinator collected

information on costs, major time expenditure, and activi-ties, including the number of patients referred over theentire project. Detailed workload measurement, receptionisttime, and questionnaires were collected twice for two-weekperiods in spring and fall 2005, in consultation with theFHNs. Methods and timing varied somewhat in each FHN.Questionnaires are available from the correspondingauthor (P.B.).

Assessment of human resources needsMean number of physicians per one full-time equivalentdietitian: This method has been used in the past to estimateservice needs (12). The number of physicians from all threeFHNs was used as the basis for calculation, without regardto patient load.

Mean number of new referrals to a full-time equivalentdietitian: The calendar data on new referrals for Januaryto December 2005 were used to calculate new referrals,defined as new persons referred for nutrition counsellingfor a new diagnosis. The patient roster as of December 2005was the denominator. An “episode of care” was defined asa person who was referred to nutrition counselling for aspecific problem, and who continued counselling forthe same problem, irrespective of follow-up timing.

The RDs collectedinformation on costs and activities over theentire project.



RESEARCH

RECHERCHE

Physician report of referral activity for one week: Allphysicians were asked to complete a yes/no checklist thatlisted all patients seen in a one-week period by initials, age,sex, referral problem, whether they had a contributingnutrition issue, and how the case was managed. Manage-ment options were a. provided nutrition/lifestyle counselling,b. referred to nurse, c. referred to RD, or d. referred to otherand, if so, whom.

Calculation of direct costs Direct costs are those that are directly incurred by having

the RD in the FHN. They include RDs’ compensation, theset-up costs (for example, purchasing weigh scales or foodmodels), and the ongoing costs that the RD requires tooperate (for example, telephonecharges, paying a booking assistant).Direct costs were calculated frommonthly data submitted to the projectcoordinator. Direct costs were trackedaccording to the category in whichthe money was spent. For the purposesof calculation, a full-time equivalent(FTE) position was assumed to equal1,950 hours of work per year.

Other direct costs could beincurred if the addition of the RD to the FHN createdsignificant additional work for the FHN staff. This wouldneed to be so significant that it would necessitate hiringadditional staff. If that were the case, another direct costwould be the cost of hiring an additional person to workat the FHN to help with the extra work created by theRD. Whether this might be necessary was assessed by thechange-in-routine questionnaire.

Calculation of indirect costsIndirect costs are not directly billable. For the purposes

of this analysis, the major indirect cost considered was thecost of communication between the RDs and other providers.For example, when the RD needed to speak with a physi-cian, she was incurring an indirect cost because the timethat the physician took to talk with her was time taken fromother work. Hence, the indirect cost of such interactionswould be the physician’s salary multiplied by the amountof time for the interaction.

To assess indirect costs, RDs completed a two-week formtwice during the study. The RDs placed a check mark inthe appropriate column for each interaction, noting withwhom the interaction took place (e.g., physician, nurse,receptionist, etc.). The RDs also indicated the averageamount of time for each interaction, and weekly interac-tion times were calculated from these data.

Sensitivity analysisA major issue in adding an RD to the team was the

potential impact on receptionist workload. As previouslymentioned, one RD made her own appointments through-out the study, while receptionists in the other two FHNsmade appointments at least part of the time. Receptionistsmaking appointments for the dietitian were asked to track this

activity for two two-week periods during the study, using acheck-off form and estimating the average call as taking twominutes.

Another aspect that was considered in the sensitivityanalysis was the possible effect of the RD on the work ofother FHN staff, as reflected in the change-in-routinequestionnaire.

Workload measurementDetailed workload assessment was carried out for two-

week periods twice during the project. The RD trackedwork time in 15-minute blocks by checking the appropriateboxes for the activity codes. Additional categories of worktime were added in this study for health promotion activities,

training, research, and travel timebetween FHN sites during a workday.

All data were analyzed using eitherMicrosoft Excel 2002 (Redmond, WA,Microsoft Corporation, 2002) or SPSS10.0 (SPSS Inc., Chicago, IL, 2000). Allaspects of the study were approved bythe Research Ethics Board at theUniversity of Guelph.

RESULTSOnly mean values are reported across the three FHNs,

both to protect the identity of individuals and to developestimates for planning purposes.

Human resources estimatesThe RDs’ paid hours ranged from 0.57 to 0.71 of an FTE,

with an average of 1,306 hours per year or 0.67 FTE. There-fore, 0.67 x 3 = 2.0 FTE RDs were working with 41 MDs. Theratio of RDs to MDs was 1:20.5.

From January to December 2005, 757 referrals weremade, according to a combined roster of 59,926 patients(as of December 2005), for an average referral rate of1.26% (range: 1.04% to 1.44%). Therefore, the 757 newreferrals translated to 379 referrals per FTE RD.

This overall referral rate includes patients who werereferred but did not book an appointment with the RD(10% of total referrals), as well as patients who booked anappointment but did not show up for the appointment (10%).Of the 757 referrals, the RDs completed baseline interviewswith 603 patients, or 302 patients per FTE RD per year.

Twenty-seven of 41 physicians (66%) completed thephysician management form. Of 1,884 patients reviewed,17.5% were reported to have a contributing nutrition prob-lem. Physicians reported that they discussed the nutritionissues with 12% of all patients and referred 25 patients, or1.3% of the patients reviewed, to the FHN RD. Another 20(1.1%) patients were referred to other community services,such as diabetes education centres.

Direct costsTwo sets of direct costs are shown in Tables 1 and 2.

Table 1 shows the actual costs to the three FHNs involvedin the project. Direct costs were based on the mean RDcontracted rate of $36.81/hour for the actual hours worked,

Of 1,884 patientsreviewed, 17.5% werereported to have acontributing nutritionproblem.

3006055DC_Research BrauerCost_S30_S38 8/9/06 1:44 PM 2

plus actual expenses, adjusted to oneyear. Table 2 shows the projected costsof adding an FTE RD to one FHN,excluding project-specific expenses.One FTE RD costs $36.81 multipliedby 1,950 hours per year for an annualcost of $71,779.50, plus expenses. Foran RD in a salaried position, vacationpay, health benefits, or remunerationin lieu of benefits might need to beconsidered. The RD might also nego-tiate other benefits, such as employer-paid continuing education expenses.

The actual expenses from the sub-mitted expense records includedmembership fees, educationalresources, supplies, printing, tele-phone, postage, courier, travel, meals,accommodation, conference costs,and help doing bookings and analyz-ing diet records. Dietitian expensesover one year were calculated as fol-lows: because data were available fora total of 19 months, all categorieswere summed and adjusted to oneyear (Table 1). Postage and courier

fees to send data to the Universityof Guelph, travel costs, meals, andaccommodation were project expensesremoved from the expense calculationsin Table 2.

Fixed costs were included in theestimated direct costs to one FHNwith one full-time RD only in Table 2,and not in the calculation of the actualcosts to the three FHNs in Table 1because the FHNs did not incur thesecosts. For this demonstration project,the RDs were expected to supplytheir own computers, which wouldnot be expected if they were hiredfull time into the FHN. Fixed costs inTable 2 would have included mostlyset-up costs, such as the cost of a desk,office chairs, and a computer for theRD; however, because they were notincurred in the demonstration pro-ject, these fixed costs were estimated.The range of fixed costs used was$2,000 to $4,000 for furniture and acomputer. The ranges are indicatedas lower and upper bounds of thedirect costs, and the middle value of$3,000 was used in the cost calcula-tion. The range reflected the widevariability in the price of these items,and allowed for the possibility thatthe FHN may already have ownedsome of these items (e.g., office fur-niture), and would not need to pur-chase them. One potential cost notincorporated into this calculation isthe possible need to rent extra spacefor the RD. The basic requirementfor an RD would be an office largeenough to seat three people comfort-ably for consultations.

The project was 19 months longand the costs were projected to twoyears. This was done for two reasons:first, to be able to show the differencein net present value (NPV) of costsversus undiscounted costs, and second,to be able to compare this with twoyears of cost projections for addingan RD to one FHN. Fixed costs,although not very large comparedwith other costs, are nonethelessincurred only in the first year. Costsare discounted at the end of the year,since salaries and expenses are paidthroughout the year, not wholly atthe beginning. The 5% discount rateis commonly used, and undiscountedcosts over the same period were also

RESEARCH

RECHERCHE


Table 1Actual direct costs for the three Family Health Networks

in the demonstration project

Compensation (3 RDs over 1 year) $144,276.79Expenses (3 RDs over 1 year) $15,893.28

Total direct costs over 1 year (including expenses)1 $160,170.07

Total direct costs over 2 years, undiscounted(includes expenses)1 $320,340.14

NPV2 of total direct costs over 2 years, discounted at 5% $297,821.90

1Expenses include membership fees, educational resources, supplies, printing, telephone,postage, couriers, travel, meals, accommodation, conferences, additional help to bookappointments and analyze food records, gifts, and equipment rental. 2NPV is the net present value of the income stream, in this case over two years.

RD = registered dietitian

Table 2Estimated direct costs for one Family Health Network with

one full-time registered dietitian

Compensation $71,779.50

Expenses (including initial fixed costs, over 1 year)1 $7,389.75

(Range: $6,389.75 to $8,389.75)

Expenses (excluding initial fixed costs, over 1 year)1 $4,389.75

Total direct costs (including expenses and fixed costs)1 $79,169.25

(Range: $78,169.25 to $80,169.25)

Total direct costs over 2 years, undiscounted $155,338.50

(Range: $154,338.50 to $156,338.50)

NPV2 of total direct costs over 2 years, discounted at 5% $144,487.04(Range: $143,534.66 to $145,439.42)

1Expenses exclude those related to the project only but do include membership fees,educational resources, supplies, printing, telephone, conferences, additional help to bookappointments and analyze food records, gifts, and equipment rental. Fixed costs include ourestimation of the cost of computer equipment and office furniture ($3,000, range: $2,000 to$4,000).2NPV is the net present value of the income stream, in this case over two years.


calculated so that these totals reflectthe range of using a discount ratebetween zero and 5%.

Indirect costsIn the indirect cost computation,

the term “others” refers to the FHNadministrator, other RDs, pharmacystaff, office staff (including nurses,receptionists, and administrative staff),and sales representatives. Because thesalaries of this mix of people varysubstantially, the average salary for alloccupations was used.

Table 3 shows the indirect projectcosts. The salaries used to calculatethe value of time are listed below thetable. The weekly averages of interac-tions are based on four weeks of datacollection: two weeks in March 2005and two weeks in October 2005.

The final calculations are the totalcosts to one FHN with one full-timeRD (Table 4). These include directand indirect costs.

Sensitivity analysis The receptionist workload was

calculated to assess how many directpatient contacts related to the RD werein each FHN. The average numberof phone calls per week for the firstset of data (collected around April2005) ranged from 11 to 51.5. Forthe second set of data (collectedaround November 2005), the rangewas 10.5 to 41.5 phone calls per weekper FHN. An increase in the numberof calls per week was seen in Stratford(from 20 to 25), while Parry Soundand Kingston both saw fewer callsduring the second collection period(from 51.5 to 41.5 and from 11 to 10.5,respectively). If each phone call tooktwo minutes (as was assumed), then51.5 calls per week represented 4.3%of a receptionist’s time.

The number of appointments madewith the RD and the number cancelledwere also recorded. Not much varia-tion occurred between the two datacollection periods in the number ofappointments kept (i.e., appointmentsmade minus appointments cancelled).The range was two to 16 appointmentsover the two-week data collectionperiod in April 2005, and five to16 appointments for two weeks inNovember 2005. However, the


RESEARCH

RECHERCHE

Table 3Average indirect costs for one Family Health Network with


Minutes of interactions Value of time(weekly average)

With physicians 80.00 $66.49With nurses 25.00 $9.73With nurse practitioners 23.75 $13.93With pharmacists 20.00 $10.80With receptionists 126.00 $27.17With others 152.50 $57.72

Total indirect costs (weekly) $185.85

Total indirect costs per year $9,664.05

Total indirect costs over 2 years, undiscounted $19,328.10

NPV1 of total indirect costs over 2 years, discounted at 5% $17,969.44

1NPV is the net present value of the income stream.

Hourly wages are based on the average annual salary of full-time workers (40 hours/week,52 weeks) for that profession in Ontario in 2000 (Statistics Canada, Earnings of Canadians,available at: www.statcan.ca), except for the salary for nurse practitioners, which is basedon the average salary advertised in job ads posted by the Registered Nursing Association ofOntario (www.rnao.org). The annual salary of general practitioners is $103,731, and hourlywages are $49.87; the annual salary of nurses (assuming they are registered) is $48,564, andhourly wages are $23.35; the annual salary of nurse practitioners is $73,200, and hourly wagesare $35.19; the annual salary of pharmacists is $67,399, and hourly wages are $32.40; the annualsalary of receptionists is $26,904, and hourly wages are $12.94; and the annual salary of otherswas taken as the average salary of all occupations, $47,232, and hourly wages of $22.71.

Table 4Estimated total direct and indirect costs for one Family Health Network with


Total direct costs per year (including fixed costs) $79,169.25

(Range: $78,169.25 to $80,169.25)

Total indirect costs per year $9,664.05

Total costs per year (first year, including fixed costs) $88,833.30

(Range: $87,833.30 to $89,833.30)

Total costs over 2 years, undiscounted $174,666.60

(Range: $173,666.60 to $175,666.60)

NPV1 of total costs over 2 years, discounted at 5% $162,456.48

(Range: $161,504.09 to $163,408.86)

1NPV is the net present value of the income stream.

Expenses include membership fees, educational resources, supplies, printing, telephone, con-ferences, hired help, gifts, and equipment rental. Fixed costs include our estimation of thecost of computer equipment and office furniture ($3,000, range: $2,000 to $4,000). Indirectcosts are those calculated in Table 3.


number of appointments cancelled was reduced; zero to14 were cancelled in the first period, and one to nine in thesecond period. The change-in-routine questionnaire did notindicate that taking calls for the RD disrupted a receptionist’swork, and the time demands (4.3%) were small.

The change-in-routine questionnaire provides additionalinsight into the value of adding an RD to a practice. Theresponse rate was 42% (47/111) for the first administrationof the questionnaire, and 56% (63/112) for the secondadministration. Cronbach’s alpha was used to measure theintercorrelation among the items in this ordinal-scale survey;it is a coefficient of internal consistency. Cronbach’s alphawas 0.879 for the eight five-level questions for the surveyconducted in spring 2005; it was 0.824 for the survey con-ducted in fall 2005. Both are considered acceptable.

The data for the two administrations were kept separateto make possible the identification of any adjustments thatthe FHN staff may have made to the presence of the RD

(Table 5). Evidence of such adjustment was seen by thesecond questionnaire administration.

Workload measurementFinally, the output for the average workload distribu-

tion for each RD at three FHNs in two time periods isillustrated in Figure 1.

DISCUSSIONCurrently, human resources planning guidelines and

costing estimates for RDs in PHC are based on very limitedinformation. Estimates developed in this study are prelimi-nary, as they depend on information from only three FHNsgathered during a demonstration project. The analysis wasfor an enhanced RD model of counselling and health pro-motion nutrition services for the first year.

We expect that these estimates will differ from otherselsewhere for several reasons. Community health centres

RESEARCH

RECHERCHE


Table 5Change-in-routine questionnaire results1

Spring 2005 (n = 47) Fall 2005 (n = 63)

% who agree Median % who agree Medianor strongly agree (range) or strongly agree (range)

General environment for change

1. The nature of my job makes structural 46 3 (2 – 5) 64 4 (1 – 5)changes easy to accommodate.

2. The climate for change is very supportive and 66 4 (2 – 5) 862 4 (2 – 5)positive in our FHN.

Attitudes and opinions about addition of the dietitian

3. Information about how the new RD would fit into our 49 3 (2 – 4) 672 4 (1 – 5)FHN was well communicated.

4. I was well prepared for my role with respect to the RD. 51 4 (1 – 4) 732 4 (1 – 5)

5. Have your duties increased as a result of the addition 33 n/a 22 n/aof the RD? (Yes/No) reporting change reporting change

5b. I did not have problems fitting the extra work that 38 3 (2 – 4) 802 4 (1 – 5)I do for the RD into my regular workload. (6/16) (24/30)

6. Having an RD in our FHN has increased my 43 3 (1 – 5) 652 4 (1 – 5)job satisfaction.

7. I feel that I can express my concerns about the new RD. 68 4 (2 – 5) 892 4 (2 – 5)

8. I feel that these concerns would be taken seriously. 62 4 (1 – 5) 842 4 (1 – 5)

Overall attitude to addition of the dietitian

9. My overall attitude toward the addition of the RD in 65 4 (2 – 4) 862 4 (1 – 4)our FHN is 1 = dissatisfied to 4 = satisfied. satisfied satisfied

1Percentage who agree or strongly agree on a five-point Likert scale, unless otherwise indicated; median and range2p<0.05 Pearson chi-square for the hypothesis that the proportion agreeing and survey administration are independent

FHN = family health network; RD = registered dietitian; n/a = not available


in Ontario serve the needs of special or high-risk popula-tions, who typically require more intensive and diverse ser-vices, such as those to promote community development(13,14). Estimates developed by private practice dietitiansmust include the costs of running a completely independentbusiness (14). Estimates developed for organizations, such asambulatory care clinics within hospitals, or regional servicessuch as the Hamilton HSO Mental Health and NutritionProgram (12), will also differ because coordinating andevaluation functions are typically centralized.

Human resources planningReferral rate estimates are based on reported case man-

agement over one week and actual referrals over one year.They were very similar – about 1.3% of FHN patients. Anadditional 1.1% were referred to other community programs,according to the case management survey. Availability ofnutrition programs varies substantially in communities, andso 1.3% to 2.4% of patients may be referred for counsellingwithin the FHN.

To calculate RD services, an estimate of the caseload thatthe RD can manage is needed. The number of patients seenby one FTE RD in this project was 379, which is lower thanthe number reported by the Hamilton HSO Mental Healthand Nutrition Program, in which one FTE RD sees 710referrals per year (12). A number of possible reasons mayexplain these differences. One factor may have been thetime required to set up new services, determine patientneeds, and establish relationships with colleagues in theFHNs and agencies. Another important difference is that thisdemonstration project included an RD evaluation compo-nent, which could not be separated from assessment andcounselling in the workload analysis, as it was embeddedwithin appointments; in addition, 11% of time was strictly

devoted to research activity. The definition of an “episodeof care” also may have differed between the programs. Suchdifferences could have a significant impact on perceivedefficiency. Overall, management of 380 new referrals a yearis a realistic estimate of the number of patients that oneFTE RD could manage in the first year of practice.

If 1.3% to 2.4% of patients require individual counsellingin a year, and a full-time RD sees 380 new referrals a year,a reasonable first estimate is that an FHN would need oneFTE RD per 15,800 to 29,000 patients ([380/0.024] to[380/0.013]) for an interdisciplinary model of nutritionservices.

As a check of the estimates developed in this analysis, theHamilton data were used to estimate the roster that oneFTE RD could manage. The demonstration project datarevealed that the average physician had 59,926/41 = 1,462or approximately 1,500 patients. The Hamilton HSO MentalHealth and Nutrition Program reported that six FTE RDsprovided services to 80 physicians (excluding administrationand evaluation), a ratio of 1 RD to 13.3 physicians (12). Whenthis number and the estimate of 1,500 patients per physi-cian are used, one FTE RD could provide services to anenrolled population of approximately 20,000 (13.3 x 1,500),an estimate that falls within the same range as the estimatesdeveloped in this project.

Numerous factors in addition to other community serviceswill affect the actual requirements. Clients’ ages, genders,health, and psychosocial characteristics, the organizationof and number of practice sites in the FHN, and thegeographic location may all have an impact. In addition,physician referral rates may vary significantly, as was foundin this study.

Despite their limits, this and other population-basedmethods for estimating RD needs will, over time, yield


RESEARCH

RECHERCHE

Figure 1Dietitian workload, averaged over the three Family Health Networks

FHN = family health network


superior estimates to those based on provider ratios, suchas RD to physician ratios. Such ratios become increasinglyproblematic as the interdisciplinary team expands andnew ways of organizing services develop.

Costing analysisDirect costs of adding an FTE RD (including expenses

and fixed costs) were calculated assuming the RD was anindependent contractor. However, if the RD were salaried,vacation pay and other benefits might have to be consid-ered. The largest component of direct costs was the RD’ssalary (approximately 80%); expenses contributed the rest.

Indirect costs of adding an RD to an FHN were incurredin this project ($9,664.05 per year [Table 3]). The commu-nication documented in this study is an integral feature ofinterdisciplinary practice, however, and the significance ofthese costs is uncertain. Such indirect costs are significant ifthe opportunity cost truly is time spent working that is notmade up in regular office hours. For example, the indirectcost that the RD incurs for physician contact is problematicif those 80 minutes per week are preventing the physicianfrom continuing to see patients or to work at the same rateas before the RD came into the FHN. The change-in-routinequestionnaire results did not indicate that the RD had pre-vented other FHN members from continuing to work at thesame pace, but data by professional designation were notcollected (Table 5).

One limitation of the costing analysis is the inability tocompare the cost estimates with the costs that would besaved by having an RD available to patients who need one,or with the advantages of having direct communicationbetween the RD and the other people involved in a patient’scare. The indirect cost savings of having an RD in an FHNare diverse and would include travel time saved by patients,the convenience of being able to make an RD appointmentat the FHN, and improved chronic disease managementthrough improved communication between the physicianand the RD. Indirect cost savings were not assessed. Furtherstudy of the indirect costs and savings of communicationin interdisciplinary practice is needed.

RELEVANCE TO PRACTICEThe advantages of having RDs in PHC settings, especially

for disease prevention and chronic disease management,are not easily calculated at this point. This is mostly becausethe advantages are long-term health benefits to patients.Diet treatment is integral to the management of diabetes,hypertension, and dyslipidemia, but effectiveness in obesitytreatment remains problematic (4). Chronic diseases are alarge burden on the Canadian health care system, however,and reducing their incidence or severity would providesubstantial benefits to the system. For example, the annualcost of obesity in Canada has been estimated to be between$1.8 billion (1997) (15) and $2.1 billion (1999) (16). Indi-rect costs could be as high as $3.7 billion (2001) (17). Giventhe potential long-term health benefits (4), and the positiveresponse to having RDs at the demonstration sites, theaddition of an RD to the interdisciplinary team in FHNs

and similar primary care organizations seems to be aworthwhile investment.

The results of this study provide some direction to groupsplanning to add RD services to their FP-based PHC practices.The study addresses a gap in the literature on humanresources planning for RD services. Additional studies arenow needed to improve this estimate by basing calculationson a wider range of PHC organizations, on a longer timeframe, and on diverse populations; in addition, effective-ness must be linked to resources. While direct costs of ser-vices are commonly estimated, receptionist and clericalcosts often are not included and should be. Informationon the indirect costs of communication and coordinationfor RD services has also been lacking. Published informa-tion on both direct and indirect costs for all models ofnutrition services in PHC settings is needed to improvethe estimates developed in this study.

AcknowledgementsThe authors gratefully acknowledge Theresa Schneider,

Deborah Northmore, and Eva West, the RDs who partici-pated in this study, for their commitment to professionalaccountability. Their dedication in completing the formsand questionnaires under the constraints of “real world”practice was critical to successful completion of the cost-ing analysis. The staff at each of the three Family HealthNetworks must also be acknowledged, as this costing analy-sis would not have been possible otherwise. The commit-ment of each lead physician – Dr. Murray Overington,Kingston FHN, Dr. Mark Wilkinson, Stratford FHN, andDr. Richard Woodhouse, Parry Sound FHN – was especiallyappreciated, as they were instrumental in ensuring successfulcompletion of the project. Dr. John Dorland of Queen’sUniversity advised on the key aspects of the costing analysis.Anne Marie Crustolo and Dr. Nicholas Kates of the HamiltonMental Health and Nutrition Program kindly shared theirclinical tracking and workload measurement tools so thatcomparable data could be generated. Two undergraduatestudents in the applied human nutrition program at theUniversity of Guelph, Catherine Shea and Susan Tran, ensuredaccurate and timely data entry. Dr. Mary Thompson of theSurvey Research Centre at the University of Waterloo reviewedthe questionnaires. Stacey Curry Gunn of Flow PublicRelations & Marketing, Guelph, ON, provided editorial assis-tance. Funded by the Ontario Primary Health Care TransitionFund, 2004-2006. This report does not represent the officialpolicy of the funding partners, the Ontario Ministry of Healthand Long-Term Care, or other organizations.

References1. Community Dietitians in Health Centres Network. Community

dietitians – trusted food and nutrition experts. Toronto: Dietitiansof Canada; 2004 [cited 2006 1 March]. Available from:http://www.dietitians.ca/pdf/CDHC_role_paper_March2004.pdf.

2. Dietitians of Canada. The role of the registered dietitian in primaryhealth care: a national perspective. Toronto: Dietitians of Canada;2001 [cited 2006 1 March]. Available from: http://www.dietitians.ca.

RESEARCH

RECHERCHE



3. American Dietetic Association. Evidence-based guides for practice.Chicago: American Dietetic Association; 2002 [cited 2006 1 March].Available from: http://www.knowledgeline.com/adaprotocols1/.

4. Ciliska D, Thomas H, Catallo C, et al. The effectiveness of nutritioninterventions for prevention and treatment of chronic diseasein primary care settings: a systematic literature review. Toronto:Dietitians of Canada; 2006 [cited 2006 20 June]. Available from:http://www.dietitians.ca/public/content/resource_centre/index.asp.

5. Green LW. What can we generalize from research on patienteducation and clinical health promotion to physician counseling ondiet? Eur J Clin Nutr 1999;53(Suppl 2):S9-18.

6. Ware JE Jr, Sherbourne DC. The MOS 36-item short-form healthsurvey (SF-36): l. Conceptual framework and item selection. Med Care1992;30:473-83.

7. Ware JE Jr, Snow KK, Kosinski M, Gandek B. SF-36 Health Surveymanual and interpretation guide. Boston: The Health Institute,New England Medical Centre; 1993.

8. Ware JE Jr, Kosinski M, Keller SD. SF-36 physical and mental healthsummary scales: a user’s manual. Boston: The Health Institute,New England Medical Centre; 1994.

9. Ware JE Jr, Kosinski M, Bayliss MS, McHorney CA, Rogers WH, Raczek A.Comparison methods for the scoring and statistical analysis of the SF-36 health profile and summary measures: summary of results fromthe medical outcomes study. Med Care 1995;33:AS264-79.

10. Performance management – Goddard Supervisor Evaluation Survey.NASA’s Goddard Space Flight Center Office of Human CapitalManagement; 2005 [cited 2006 16 March]. Available from:http://ohcm.gsfc.nasa.gov/home.htm.

11. Day M. Step by step guide to employee satisfaction surveys.SearchWarp.com. 18-10-2005 [cited 2006 16 March]. Available from:http://searchwarp.com/swa21348.htm.

12. Crustolo AM, Kates N, Ackerman S, Schamehorn S. Integratingnutrition services into primary care experience in Hamilton, Ont. Can Fam Physician 2005;51:1647-53.

13. Davidson B, Dietrich L, Brauer P. Key informant interviews: dietitianservices in current programs 2005. Toronto: Dietitians of Canada; 2006[cited 2006 20 June]. Available from: http://www.dietitians.ca/public/content/resource_centre/index.asp.

14. Davison K, Mor A, Charlebois H. What are entrepreneurial dietitianscharging? The Consulting Dietitians Network National Fee Survey.Can J Diet Prac Res 2004;65:186-90.

15. Birmingham CL, Muller JL, Palepu A, Spinelli JJ, Anis AH. The cost ofobesity in Canada. CMAJ 1999;160:483-8.

16. Katzmarzyk PT, Gledhill N, Shephard RJ. The economic burden ofphysical inactivity in Canada. CMAJ 2000;163:1435-40.

17. Katzmarzyk PT, Janssen I. The economic costs associated with physicalinactivity and obesity in Canada: an update. Can J Appl Physiol2004;29:90-115.


RESEARCH

RECHERCHE

REGIONAL EVENTSDietitians of Canada offers regional workshopstailored to the interests of dietitians and other health professionals.COMING SOONFood Allergies: What’s new in clinical, community,school and food service applications

• Thunder Bay – September 13, 2006• Saskatoon – September 15, 2006• Winnipeg – September 16, 2006

Atlantic Fall Conference 2006: Back to and Beyondthe Basics – September 29-30, 2006, Moncton

CSO Regional Conference: Explore the Diversity ofPractice -- November 10, 2006, Toronto

Making Adult Education Work in Health Care –November 14, 2006, Winnipeg

For details on how to register, visit the DC web site at www.dietitians.ca/public/content/resource_centre/pd_events.asp

DIETETICS @ WORKA practical, online professional support servicefeaturing courses that present new science inprofessional practice terms – all from the convenience of your office or home.

DIETARY SUPPLEMENTS• Lesson 1 – Vitamin/Mineral Supplements• Lesson 2 – Herbal Supplements – is our most

recent online offering.• Lesson 3 – Sports Supplements – will be

available in early 2007. Authors: Susie Langleyand Kelly Anne Erdman

DIETARY REFERENCE INTAKESHave you completed the first 6 lessons of the DRIcourse? If not, you will want to get caught up andbe ready when Lessons 7 and 8 of the DietaryReference Intakes course debut in early 2007.Author: Dr. Susan Barr

Visit www.dieteticsatwork.com for registration andpricing details.

INVEST IN YOURSELF – DC CAN HELP


phc estimation human resource needs

Health & Medicine